Billboard Top10: Top10 classification:: template2
bdanalytics
Date: (Sun) Jun 28, 2015
Introduction:
Data: Source: Training: https://courses.edx.org/asset-v1:MITx+15.071x_2a+2T2015+type@asset+block/songs.csv
New:
Time period:
Synopsis:
Based on analysis utilizing <> techniques,
Regression results: First run:
Classification results: First run:
Prediction Accuracy Enhancement Options:
- import.data chunk:
- which obs should be in fit vs. OOB (currently dirty.0 vs .1 is split 50%)
- inspect.data chunk:
- For date variables
- Appropriate factors ?
- Different / More last* features ?
- For date variables
- scrub.data chunk:
- transform.data chunk:
- derive features from multiple features
- manage.missing.data chunk:
- Not fill missing vars
- Fill missing numerics with a different algorithm
- Fill missing chars with data based on clusters
- extract.features chunk:
- Text variables: move to date extraction chunk ???
- Mine acronyms
- Mine places
- Text variables: move to date extraction chunk ???
- Review set_global_options chunk after features are finalized
[](.png)
Potential next steps include:
- Organization:
- Categorize by chunk
- Priority criteria:
- Ease of change
- Impacts report
- Cleans innards
- Bug report
- all chunks:
- at chunk-end rm(!glb_)
- manage.missing.data chunk:
- cleaner way to manage re-splitting of training vs. new entity
- extract.features chunk:
- Add n-grams for glb_txt_vars
- “RTextTools”, “tau”, “RWeka”, and “textcat” packages
- Convert user-specified mutate code to config specs
- Add n-grams for glb_txt_vars
- fit.models chunk:
- Prediction accuracy scatter graph:
- Add tiles (raw vs. PCA)
- Use shiny for drop-down of “important” features
Use plot.ly for interactive plots ?
- Change .fit suffix of model metrics to .mdl if it’s data independent (e.g. AIC, Adj.R.Squared - is it truly data independent ?, etc.)
- move model_type parameter to myfit_mdl before indep_vars_vctr (keep all model_* together)
- create a custom model for rpart that has minbucket as a tuning parameter
varImp for randomForest crashes in caret version:6.0.41 -> submit bug report
- Probability handling for multinomials vs. desired binomial outcome
- ROCR currently supports only evaluation of binary classification tasks (version 1.0.7)
extensions toward multiclass classification are scheduled for the next release
- Skip trControl.method=“cv” for dummy classifier ?
- Add custom model to caret for a dummy (baseline) classifier (binomial & multinomial) that generates proba/outcomes which mimics the freq distribution of glb_rsp_var values; Right now glb_dmy_glm_mdl always generates most frequent outcome in training data
glm_dmy_mdl should use the same method as glm_sel_mdl until custom dummy classifer is implemented
- fit.all.training chunk:
- myplot_prediction_classification: displays ‘x’ instead of ‘+’ when there are no prediction errors
- Compare glb_sel_mdl vs. glb_fin_mdl:
- varImp
- Prediction differences (shd be minimal ?)
- Move glb_analytics_diag_plots to mydsutils.R: (+) Easier to debug (-) Too many glb vars used
- Add print(ggplot.petrinet(glb_analytics_pn) + coord_flip()) at the end of every major chunk
- Parameterize glb_analytics_pn
- Move glb_impute_missing_data to mydsutils.R: (-) Too many glb vars used; glb_<>_df reassigned
- Replicate myfit_mdl_classification features in myfit_mdl_regression
- Do non-glm methods handle interaction terms ?
- f-score computation for classifiers should be summation across outcomes (not just the desired one ?)
- Add accuracy computation to glb_dmy_mdl in predict.data.new chunk
- Why does splitting fit.data.training.all chunk into separate chunks add an overhead of ~30 secs ? It’s not rbind b/c other chunks have lower elapsed time. Is it the number of plots ?
- Incorporate code chunks in print_sessionInfo
- Test against
- projects in github.com/bdanalytics
- lectures in jhu-datascience track
Analysis:
rm(list=ls())
set.seed(12345)
options(stringsAsFactors=FALSE)
source("~/Dropbox/datascience/R/myscript.R")
source("~/Dropbox/datascience/R/mydsutils.R")## Loading required package: caret
## Loading required package: lattice
## Loading required package: ggplot2source("~/Dropbox/datascience/R/myplot.R")
source("~/Dropbox/datascience/R/mypetrinet.R")
source("~/Dropbox/datascience/R/myplclust.R")
# Gather all package requirements here
suppressPackageStartupMessages(require(doMC))
registerDoMC(4) # max(length(glb_txt_vars), glb_n_cv_folds) + 1
#packageVersion("snow")
#require(sos); findFn("cosine", maxPages=2, sortby="MaxScore")
# Analysis control global variables
glb_trnng_url <- "https://courses.edx.org/asset-v1:MITx+15.071x_2a+2T2015+type@asset+block/songs.csv"
glb_newdt_url <- "<newdt_url>"
glb_out_pfx <- "template2_"
glb_save_envir <- FALSE # or TRUE
glb_is_separate_newobs_dataset <- FALSE # or TRUE
glb_split_entity_newobs_datasets <- TRUE # or FALSE
glb_split_newdata_method <- "condition" # "condition" or "sample" or "copy"
glb_split_newdata_condition <- "year >= 2010"
glb_split_newdata_size_ratio <- 0.3 # > 0 & < 1
glb_split_sample.seed <- 123 # or any integer
glb_max_fitobs <- NULL # or any integer
glb_is_regression <- FALSE; glb_is_classification <- !glb_is_regression;
glb_is_binomial <- TRUE # or TRUE or FALSE
glb_rsp_var_raw <- "Top10"
# for classification, the response variable has to be a factor
glb_rsp_var <- "Top10.fctr"
# if the response factor is based on numbers/logicals e.g (0/1 OR TRUE/FALSE vs. "A"/"B"),
# or contains spaces (e.g. "Not in Labor Force")
# caret predict(..., type="prob") crashes
glb_map_rsp_raw_to_var <- function(raw) {
# return(log(raw))
ret_vals <- rep_len(NA, length(raw)); ret_vals[!is.na(raw)] <- ifelse(raw[!is.na(raw)] == 1, "Y", "N"); return(relevel(as.factor(ret_vals), ref="N"))
# #as.factor(paste0("B", raw))
# #as.factor(gsub(" ", "\\.", raw))
}
glb_map_rsp_raw_to_var(c(1, 1, 0, 0, NA))## [1] Y Y N N <NA>
## Levels: N Yglb_map_rsp_var_to_raw <- function(var) {
# return(exp(var))
as.numeric(var) - 1
# #as.numeric(var)
# #gsub("\\.", " ", levels(var)[as.numeric(var)])
# c("<=50K", " >50K")[as.numeric(var)]
# #c(FALSE, TRUE)[as.numeric(var)]
}
glb_map_rsp_var_to_raw(glb_map_rsp_raw_to_var(c(1, 1, 0, 0, NA)))## [1] 1 1 0 0 NAif ((glb_rsp_var != glb_rsp_var_raw) & is.null(glb_map_rsp_raw_to_var))
stop("glb_map_rsp_raw_to_var function expected")
glb_rsp_var_out <- paste0(glb_rsp_var, ".predict.") # model_id is appended later
# List info gathered for various columns
# <col_name>: <description>; <notes>
# year = the year the song was released
# songtitle = the title of the song
# artistname = the name of the artist of the song
# songID and artistID = identifying variables for the song and artist
# timesignature and timesignature_confidence = a variable estimating the time signature of the song, and the confidence in the estimate
# loudness = a continuous variable indicating the average amplitude of the audio in decibels
# tempo and tempo_confidence = a variable indicating the estimated beats per minute of the song, and the confidence in the estimate
# key and key_confidence = a variable with twelve levels indicating the estimated key of the song (C, C#, . . ., B), and the confidence in the estimate
# energy = a variable that represents the overall acoustic energy of the song, using a mix of features such as loudness
# pitch = a continuous variable that indicates the pitch of the song
# timbre_0_min, timbre_0_max, timbre_1_min, timbre_1_max, . . . , timbre_11_min, and timbre_11_max = variables that indicate the minimum/maximum values over all segments for each of the twelve values in the timbre vector (resulting in 24 continuous variables)
# Top10 = a binary variable indicating whether or not the song made it to the Top 10 of the Billboard Hot 100 Chart (1 if it was in the top 10, and 0 if it was not)
# If multiple vars are parts of id, consider concatenating them to create one id var
# If glb_id_var == NULL, ".rownames <- row.names()" is the default
glb_id_var <- NULL #c("songID")
glb_category_vars <- NULL # or c("<var1>", "<var2>")
glb_drop_vars <- c(NULL) # or c("<col_name>")
glb_map_vars <- NULL # or c("<var1>", "<var2>")
glb_map_urls <- list();
# glb_map_urls[["<var1>"]] <- "<var1.url>"
glb_assign_pairs_lst <- NULL;
# glb_assign_pairs_lst[["<var1>"]] <- list(from=c(NA),
# to=c("NA.my"))
glb_assign_vars <- names(glb_assign_pairs_lst)
# Derived features
glb_derive_lst <- NULL;
# Add logs of numerics that are not distributed normally -> do automatically ???
# glb_derive_lst[["<feat.sfx>"]] <- list(
# mapfn=function(Rasmussen) { return(ifelse(sign(Rasmussen) >= 0, 1, 0)) }
# mapfn=function(PropR) { return(as.factor(ifelse(PropR >= 0.5, "Y", "N"))) }
# mapfn=function(Week) { return(substr(Week, 1, 10)) }
# mapfn=function(raw) { tfr_raw <- as.character(cut(raw, 5));
# tfr_raw[is.na(tfr_raw)] <- "NA.my";
# return(as.factor(tfr_raw)) }
# , args=c("raw"))
# mapfn=function(PTS, oppPTS) { return(PTS - oppPTS) }
# , args=c("PTS", "oppPTS"))
# # If glb_allobs_df is not sorted in the desired manner
# mapfn=function(Week) { return(coredata(lag(zoo(orderBy(~Week, glb_allobs_df)$ILI), -2, na.pad=TRUE))) }
# mapfn=function(ILI) { return(coredata(lag(zoo(ILI), -2, na.pad=TRUE))) }
# mapfn=function(ILI.2.lag) { return(log(ILI.2.lag)) }
# glb_derive_lst[["<txt_var>.niso8859.log"]] <- list(
# mapfn=function(<txt_var>) { match_lst <- gregexpr("&#[[:digit:]]{3};", <txt_var>)
# match_num_vctr <- unlist(lapply(match_lst,
# function(elem) length(elem)))
# return(log(1 + match_num_vctr)) }
# , args=c("<txt_var>"))
# mapfn=function(raw) { mod_raw <- raw;
# mod_raw <- gsub("&#[[:digit:]]{3};", " ", mod_raw);
# # Modifications for this exercise only
# mod_raw <- gsub("\\bgoodIn ", "good In", mod_raw);
# return(mod_raw)
# # Create user-specified pattern vectors
# #sum(mycount_pattern_occ("Metropolitan Diary:", glb_allobs_df$Abstract) > 0)
# if (txt_var %in% c("Snippet", "Abstract")) {
# txt_X_df[, paste0(txt_var_pfx, ".P.metropolitan.diary.colon")] <-
# as.integer(0 + mycount_pattern_occ("Metropolitan Diary:",
# glb_allobs_df[, txt_var]))
#summary(glb_allobs_df[ ,grep("P.on.this.day", names(glb_allobs_df), value=TRUE)])
# glb_derive_lst[["<var1>"]] <- glb_derive_lst[["<var2>"]]
glb_derive_vars <- names(glb_derive_lst)
# tst <- "PropR.fctr"; args_lst <- NULL; for (arg in glb_derive_lst[[tst]]$args) args_lst[[arg]] <- glb_allobs_df[, arg]; print(head(args_lst[[arg]])); print(head(drv_vals <- do.call(glb_derive_lst[[tst]]$mapfn, args_lst)));
# print(which_ix <- which(args_lst[[arg]] == 0.75)); print(drv_vals[which_ix]);
glb_date_vars <- NULL # or c("<date_var>")
glb_date_fmts <- list(); #glb_date_fmts[["<date_var>"]] <- "%m/%e/%y"
glb_date_tzs <- list(); #glb_date_tzs[["<date_var>"]] <- "America/New_York"
#grep("America/New", OlsonNames(), value=TRUE)
glb_txt_vars <- NULL # or c("<txt_var1>", "<txt_var2>")
#Sys.setlocale("LC_ALL", "C") # For english
glb_append_stop_words <- list()
# Remember to use unstemmed words
#orderBy(~ -cor.y.abs, subset(glb_feats_df, grepl("[HSA]\\.T\\.", id) & !is.na(cor.high.X)))
#dsp_obs(Headline.contains="polit")
#subset(glb_allobs_df, H.T.compani > 0)[, c("UniqueID", "Headline", "H.T.compani")]
# glb_append_stop_words[["<txt_var1>"]] <- c(NULL
# # ,"<word1>" # <reason1>
# )
#subset(glb_allobs_df, S.T.newyorktim > 0)[, c("UniqueID", "Snippet", "S.T.newyorktim")]
#glb_txt_lst[["Snippet"]][which(glb_allobs_df$UniqueID %in% c(8394, 8317, 8339, 8350, 8307))]
glb_important_terms <- list()
# Remember to use stemmed terms
glb_sprs_thresholds <- NULL # or c(0.988, 0.970, 0.970) # Generates 29, 22, 22 terms
# Properties:
# numrows(glb_feats_df) << numrows(glb_fitobs_df)
# Select terms that appear in at least 0.2 * O(FP/FN(glb_OOBobs_df))
# numrows(glb_OOBobs_df) = 1.1 * numrows(glb_newobs_df)
names(glb_sprs_thresholds) <- glb_txt_vars
# User-specified exclusions
glb_exclude_vars_as_features <- c("songID", "songtitle", "artistID", "artistname", "year")
if (glb_rsp_var_raw != glb_rsp_var)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_rsp_var_raw)
# List feats that shd be excluded due to known causation by prediction variable
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c(NULL)) # or c("<col_name>")
glb_impute_na_data <- FALSE # or TRUE
glb_mice_complete.seed <- 144 # or any integer
glb_cluster <- FALSE # or TRUE
glb_interaction_only_features <- NULL # or ???
glb_models_lst <- list(); glb_models_df <- data.frame()
# Regression
if (glb_is_regression)
glb_models_method_vctr <- c("lm", "glm", "bayesglm", "rpart", "rf") else
# Classification
if (glb_is_binomial)
glb_models_method_vctr <- c("glm", "bayesglm", "rpart", "rf") else
glb_models_method_vctr <- c("rpart", "rf")
# Baseline prediction model feature(s)
glb_Baseline_mdl_var <- NULL # or c("<col_name>")
glb_model_metric_terms <- NULL # or matrix(c(
# 0,1,2,3,4,
# 2,0,1,2,3,
# 4,2,0,1,2,
# 6,4,2,0,1,
# 8,6,4,2,0
# ), byrow=TRUE, nrow=5)
glb_model_metric <- NULL # or "<metric_name>"
glb_model_metric_maximize <- NULL # or FALSE (TRUE is not the default for both classification & regression)
glb_model_metric_smmry <- NULL # or function(data, lev=NULL, model=NULL) {
# confusion_mtrx <- t(as.matrix(confusionMatrix(data$pred, data$obs)))
# #print(confusion_mtrx)
# #print(confusion_mtrx * glb_model_metric_terms)
# metric <- sum(confusion_mtrx * glb_model_metric_terms) / nrow(data)
# names(metric) <- glb_model_metric
# return(metric)
# }
glb_tune_models_df <-
rbind(
#data.frame(parameter="cp", min=0.00005, max=0.00005, by=0.000005),
#seq(from=0.01, to=0.01, by=0.01)
#data.frame(parameter="mtry", min=080, max=100, by=10),
#data.frame(parameter="mtry", min=08, max=10, by=1),
data.frame(parameter="dummy", min=2, max=4, by=1)
)
# or NULL
glb_n_cv_folds <- 3 # or NULL
glb_clf_proba_threshold <- NULL # 0.5
# Model selection criteria
if (glb_is_regression)
glb_model_evl_criteria <- c("min.RMSE.OOB", "max.R.sq.OOB", "max.Adj.R.sq.fit")
if (glb_is_classification) {
if (glb_is_binomial)
glb_model_evl_criteria <-
c("max.Accuracy.OOB", "max.auc.OOB", "max.Kappa.OOB", "min.aic.fit") else
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB")
}
glb_sel_mdl_id <- NULL # or "<model_id_prefix>.<model_method>"
glb_fin_mdl_id <- glb_sel_mdl_id # or "Final"
# Depict process
glb_analytics_pn <- petrinet(name="glb_analytics_pn",
trans_df=data.frame(id=1:6,
name=c("data.training.all","data.new",
"model.selected","model.final",
"data.training.all.prediction","data.new.prediction"),
x=c( -5,-5,-15,-25,-25,-35),
y=c( -5, 5, 0, 0, -5, 5)
),
places_df=data.frame(id=1:4,
name=c("bgn","fit.data.training.all","predict.data.new","end"),
x=c( -0, -20, -30, -40),
y=c( 0, 0, 0, 0),
M0=c( 3, 0, 0, 0)
),
arcs_df=data.frame(
begin=c("bgn","bgn","bgn",
"data.training.all","model.selected","fit.data.training.all",
"fit.data.training.all","model.final",
"data.new","predict.data.new",
"data.training.all.prediction","data.new.prediction"),
end =c("data.training.all","data.new","model.selected",
"fit.data.training.all","fit.data.training.all","model.final",
"data.training.all.prediction","predict.data.new",
"predict.data.new","data.new.prediction",
"end","end")
))
#print(ggplot.petrinet(glb_analytics_pn))
print(ggplot.petrinet(glb_analytics_pn) + coord_flip())## Loading required package: grid
glb_analytics_avl_objs <- NULL
glb_chunks_df <- myadd_chunk(NULL, "import.data")## label step_major step_minor bgn end elapsed
## 1 import.data 1 0 8.268 NA NAStep 1.0: import data
chunk option: eval=
#glb_chunks_df <- myadd_chunk(NULL, "import.data")
glb_trnobs_df <- myimport_data(url=glb_trnng_url, comment="glb_trnobs_df",
force_header=TRUE)## [1] "Reading file ./data/songs.csv..."
## [1] "dimensions of data in ./data/songs.csv: 7,574 rows x 39 cols"
## year songtitle artistname
## 1 2010 This Is the House That Doubt Built A Day to Remember
## 2 2010 Sticks & Bricks A Day to Remember
## 3 2010 All I Want A Day to Remember
## 4 2010 It's Complicated A Day to Remember
## 5 2010 2nd Sucks A Day to Remember
## 6 2010 Better Off This Way A Day to Remember
## songID artistID timesignature
## 1 SOBGGAB12C5664F054 AROBSHL1187B9AFB01 3
## 2 SOPAQHU1315CD47F31 AROBSHL1187B9AFB01 4
## 3 SOOIZOU1376E7C6386 AROBSHL1187B9AFB01 4
## 4 SODRYWD1315CD49DBE AROBSHL1187B9AFB01 4
## 5 SOICMQB1315CD46EE3 AROBSHL1187B9AFB01 4
## 6 SOCEYON1315CD4A23E AROBSHL1187B9AFB01 4
## timesignature_confidence loudness tempo tempo_confidence key
## 1 0.853 -4.262 91.525 0.953 11
## 2 1.000 -4.051 140.048 0.921 10
## 3 1.000 -3.571 160.512 0.489 2
## 4 1.000 -3.815 97.525 0.794 1
## 5 0.788 -4.707 140.053 0.286 6
## 6 1.000 -3.807 160.366 0.347 4
## key_confidence energy pitch timbre_0_min timbre_0_max timbre_1_min
## 1 0.453 0.9666556 0.024 0.002 57.342 -6.496
## 2 0.469 0.9847095 0.025 0.000 57.414 -37.351
## 3 0.209 0.9899004 0.026 0.003 57.422 -17.222
## 4 0.632 0.9392072 0.013 0.000 57.765 -32.083
## 5 0.483 0.9877376 0.063 0.000 56.872 -223.922
## 6 0.627 0.9799530 0.038 0.000 57.083 -40.408
## timbre_1_max timbre_2_min timbre_2_max timbre_3_min timbre_3_max
## 1 171.093 -81.664 95.117 -285.049 259.426
## 2 171.130 -149.589 180.334 -380.143 384.166
## 3 171.060 -72.912 157.925 -203.984 251.258
## 4 220.895 -138.596 173.365 -73.490 373.492
## 5 171.130 -147.153 166.008 -128.082 389.419
## 6 174.002 -83.829 126.879 -100.119 173.607
## timbre_4_min timbre_4_max timbre_5_min timbre_5_max timbre_6_min
## 1 -40.385 73.630 -104.683 183.089 -88.771
## 2 -48.662 100.414 -87.267 42.798 -86.895
## 3 -66.044 152.095 -98.673 141.365 -88.874
## 4 -55.607 119.158 -77.515 141.178 -70.790
## 5 -43.908 99.332 -96.147 38.303 -110.757
## 6 -33.789 66.904 -84.451 47.268 -71.219
## timbre_6_max timbre_7_min timbre_7_max timbre_8_min timbre_8_max
## 1 73.549 -71.127 82.475 -52.025 39.116
## 2 75.455 -65.807 106.918 -61.320 35.378
## 3 66.504 -67.433 80.621 -59.773 45.979
## 4 64.540 -63.667 96.675 -78.660 41.088
## 5 72.391 -55.935 110.332 -56.450 37.555
## 6 71.239 -79.948 91.117 -54.378 53.808
## timbre_9_min timbre_9_max timbre_10_min timbre_10_max timbre_11_min
## 1 -35.368 71.642 -126.440 18.658 -44.770
## 2 -81.928 74.574 -103.808 121.935 -38.892
## 3 -46.293 59.904 -108.313 33.300 -43.733
## 4 -49.194 95.440 -102.676 46.422 -59.439
## 5 -48.588 67.570 -52.796 22.888 -50.414
## 6 -33.183 54.657 -64.478 34.522 -40.922
## timbre_11_max Top10
## 1 25.989 0
## 2 22.513 0
## 3 25.744 0
## 4 37.082 0
## 5 32.758 0
## 6 36.453 0
## year songtitle artistname songID
## 262 2010 When Love Gets a Hold of You Reba SOCZKPW12C5641317E
## 1154 2008 Take You There Sean Kingston SORKRBM1374015BCDA
## 3857 2002 One Faith Hill SOSOGQP135C392FC94
## 5511 1998 Domestic Violence RZA SOCZFNY1311AFE8F95
## 5569 1998 Feel This Way Zebrahead SOGZAIO13129A92246
## 7495 1990 Liberty Ship The La's SOWDRVI12A8C13C63C
## artistID timesignature timesignature_confidence loudness
## 262 ARPH1W61187B9B4CD8 4 0.655 -4.984
## 1154 AR9LJ9D1187FB4303D 4 1.000 -12.654
## 3857 ARVNGA71187FB3C107 4 1.000 -3.493
## 5511 AROUGGS1187FB51CA1 4 0.937 -7.731
## 5569 ARCBQRB1187FB52F88 4 1.000 -6.590
## 7495 AR8V85J1187B9ACB4A 4 1.000 -8.529
## tempo tempo_confidence key key_confidence energy pitch
## 262 128.090 1.000 0 0.660 0.7772171 0.020
## 1154 116.024 0.419 8 0.237 0.6961294 0.009
## 3857 79.001 0.601 6 0.254 0.7157751 0.005
## 5511 90.840 0.754 1 0.060 0.6650303 0.003
## 5569 111.221 1.000 6 0.295 0.9551664 0.004
## 7495 99.405 0.820 11 0.019 0.7127939 0.030
## timbre_0_min timbre_0_max timbre_1_min timbre_1_max timbre_2_min
## 262 0.020 58.303 -129.068 170.769 -124.348
## 1154 34.605 50.296 -42.595 233.750 -142.882
## 3857 0.000 59.939 -132.546 243.926 -193.551
## 5511 0.000 54.832 -64.913 243.222 -159.325
## 5569 0.000 55.962 -238.622 180.486 -127.988
## 7495 0.006 55.901 -51.119 230.989 -117.601
## timbre_2_max timbre_3_min timbre_3_max timbre_4_min timbre_4_max
## 262 174.242 -157.578 95.009 -56.617 79.562
## 1154 127.075 -132.503 96.467 -66.977 118.690
## 3857 167.289 -273.823 183.082 -57.079 165.864
## 5511 142.525 -156.074 319.301 -119.629 119.534
## 5569 122.048 -102.617 93.045 -71.845 113.590
## 7495 100.957 -171.453 95.348 -106.919 57.326
## timbre_5_min timbre_5_max timbre_6_min timbre_6_max timbre_7_min
## 262 -85.451 57.943 -77.515 61.073 -67.306
## 1154 -74.795 110.296 -90.218 63.028 -103.846
## 3857 -112.776 199.830 -105.671 79.788 -105.861
## 5511 -129.610 274.626 -102.462 56.879 -124.531
## 5569 -123.461 45.145 -84.796 97.613 -82.657
## 7495 -82.783 185.709 -106.765 58.805 -80.145
## timbre_7_max timbre_8_min timbre_8_max timbre_9_min timbre_9_max
## 262 59.944 -83.330 44.507 -45.331 49.539
## 1154 94.827 -63.906 39.722 -49.856 74.920
## 3857 95.457 -70.141 75.744 -64.718 68.458
## 5511 99.836 -81.727 74.712 -72.099 84.240
## 5569 65.853 -45.191 80.223 -81.819 73.628
## 7495 55.620 -98.838 30.321 -55.246 55.172
## timbre_10_min timbre_10_max timbre_11_min timbre_11_max Top10
## 262 -83.405 50.168 -59.199 33.327 0
## 1154 -103.825 82.409 -63.028 42.040 1
## 3857 -99.661 72.336 -67.201 65.759 0
## 5511 -97.461 85.892 -59.753 56.363 0
## 5569 -76.362 49.676 -71.164 35.919 0
## 7495 -50.047 32.024 -36.554 63.253 0
## year songtitle artistname songID
## 7569 1990 Red Hot & Ready Y&T SOHVWXR12A6D4FC59C
## 7570 1990 She's Gone Y&T SOSIEQB12A6D4FC59D
## 7571 1990 Let It Out Y&T SOIGIQI12A6D4FC59E
## 7572 1990 Ten Lovers Y&T SOLNEQO12A6D4FC59F
## 7573 1990 Goin' Off The Deep End Y&T SOEMJEP12A58A7E7B5
## 7574 1990 Surrender Y&T SOLBQSX12A6D4FC5A0
## artistID timesignature timesignature_confidence loudness
## 7569 ARGQANQ11F50C4769E 4 1.000 -10.574
## 7570 ARGQANQ11F50C4769E 4 1.000 -10.197
## 7571 ARGQANQ11F50C4769E 4 1.000 -12.392
## 7572 ARGQANQ11F50C4769E 4 0.984 -10.304
## 7573 ARGQANQ11F50C4769E 4 0.907 -9.295
## 7574 ARGQANQ11F50C4769E 4 0.987 -9.762
## tempo tempo_confidence key key_confidence energy pitch
## 7569 137.872 1.000 2 0.006 0.9736458 0.030
## 7570 93.140 0.859 5 0.889 0.9429917 0.016
## 7571 79.858 0.196 9 0.149 0.8124216 0.012
## 7572 91.760 0.592 2 0.077 0.7368709 0.016
## 7573 110.907 0.838 9 0.621 0.9900532 0.061
## 7574 139.650 0.781 6 0.193 0.9448227 0.027
## timbre_0_min timbre_0_max timbre_1_min timbre_1_max timbre_2_min
## 7569 0 51.847 -18.801 203.289 -141.762
## 7570 0 53.462 -22.878 202.424 -77.467
## 7571 0 51.354 -91.916 202.639 -76.736
## 7572 0 53.358 -10.087 202.877 -48.768
## 7573 0 52.928 -15.289 175.845 -119.044
## 7574 0 52.989 -32.907 195.601 -146.973
## timbre_2_max timbre_3_min timbre_3_max timbre_4_min timbre_4_max
## 7569 157.204 -120.035 364.391 -53.240 99.468
## 7570 174.663 -170.802 201.677 -45.807 109.383
## 7571 165.643 -186.653 166.200 -63.367 132.495
## 7572 194.665 -201.008 132.174 -55.916 129.674
## 7573 164.155 -162.944 391.622 -42.728 137.159
## 7574 150.596 -79.356 307.651 -39.215 68.426
## timbre_5_min timbre_5_max timbre_6_min timbre_6_max timbre_7_min
## 7569 -82.554 115.220 -91.407 62.922 -57.069
## 7570 -75.011 103.305 -130.215 59.775 -61.197
## 7571 -115.231 86.509 -83.905 102.373 -66.416
## 7572 -82.989 166.003 -83.246 62.951 -69.512
## 7573 -80.171 92.551 -64.419 74.428 -38.794
## 7574 -77.927 202.200 -55.617 98.615 -71.984
## timbre_7_max timbre_8_min timbre_8_max timbre_9_min timbre_9_max
## 7569 96.283 -56.468 34.205 -50.403 51.369
## 7570 38.120 -66.638 50.105 -40.692 43.763
## 7571 83.454 -83.707 44.124 -51.628 69.929
## 7572 103.413 -101.464 36.152 -45.387 48.352
## 7573 108.688 -55.893 42.222 -76.631 68.336
## 7574 87.098 -48.440 52.198 -60.673 48.418
## timbre_10_min timbre_10_max timbre_11_min timbre_11_max Top10
## 7569 -63.965 82.665 -47.429 58.158 0
## 7570 -59.707 49.414 -53.970 68.303 0
## 7571 -97.153 36.745 -61.243 56.902 0
## 7572 -57.103 67.641 -53.729 65.176 0
## 7573 -83.284 56.476 -51.687 59.427 0
## 7574 -120.625 49.593 -47.656 70.005 0
## 'data.frame': 7574 obs. of 20 variables:
## $ year : int 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 ...
## $ songtitle : chr "This Is the House That Doubt Built" "Sticks & Bricks" "All I Want" "It's Complicated" ...
## $ artistname : chr "A Day to Remember" "A Day to Remember" "A Day to Remember" "A Day to Remember" ...
## $ songID : chr "SOBGGAB12C5664F054" "SOPAQHU1315CD47F31" "SOOIZOU1376E7C6386" "SODRYWD1315CD49DBE" ...
## $ artistID : chr "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" ...
## $ timesignature : int 3 4 4 4 4 4 4 4 4 4 ...
## $ timesignature_confidence: num 0.853 1 1 1 0.788 1 0.968 0.861 0.622 0.938 ...
## $ loudness : num -4.26 -4.05 -3.57 -3.81 -4.71 ...
## $ tempo : num 91.5 140 160.5 97.5 140.1 ...
## $ tempo_confidence : num 0.953 0.921 0.489 0.794 0.286 0.347 0.273 0.83 0.018 0.929 ...
## $ key : int 11 10 2 1 6 4 10 5 9 11 ...
## $ key_confidence : num 0.453 0.469 0.209 0.632 0.483 0.627 0.715 0.423 0.751 0.602 ...
## $ energy : num 0.967 0.985 0.99 0.939 0.988 ...
## $ pitch : num 0.024 0.025 0.026 0.013 0.063 0.038 0.026 0.033 0.027 0.004 ...
## $ timbre_0_min : num 0.002 0 0.003 0 0 ...
## $ timbre_0_max : num 57.3 57.4 57.4 57.8 56.9 ...
## $ timbre_1_min : num -6.5 -37.4 -17.2 -32.1 -223.9 ...
## $ timbre_1_max : num 171 171 171 221 171 ...
## $ timbre_2_min : num -81.7 -149.6 -72.9 -138.6 -147.2 ...
## $ timbre_2_max : num 95.1 180.3 157.9 173.4 166 ...
## NULL
## 'data.frame': 7574 obs. of 21 variables:
## $ timbre_2_min : num -81.7 -149.6 -72.9 -138.6 -147.2 ...
## $ timbre_2_max : num 95.1 180.3 157.9 173.4 166 ...
## $ timbre_3_min : num -285 -380.1 -204 -73.5 -128.1 ...
## $ timbre_3_max : num 259 384 251 373 389 ...
## $ timbre_4_min : num -40.4 -48.7 -66 -55.6 -43.9 ...
## $ timbre_4_max : num 73.6 100.4 152.1 119.2 99.3 ...
## $ timbre_5_min : num -104.7 -87.3 -98.7 -77.5 -96.1 ...
## $ timbre_5_max : num 183.1 42.8 141.4 141.2 38.3 ...
## $ timbre_6_min : num -88.8 -86.9 -88.9 -70.8 -110.8 ...
## $ timbre_6_max : num 73.5 75.5 66.5 64.5 72.4 ...
## $ timbre_7_min : num -71.1 -65.8 -67.4 -63.7 -55.9 ...
## $ timbre_7_max : num 82.5 106.9 80.6 96.7 110.3 ...
## $ timbre_8_min : num -52 -61.3 -59.8 -78.7 -56.5 ...
## $ timbre_8_max : num 39.1 35.4 46 41.1 37.6 ...
## $ timbre_9_min : num -35.4 -81.9 -46.3 -49.2 -48.6 ...
## $ timbre_9_max : num 71.6 74.6 59.9 95.4 67.6 ...
## $ timbre_10_min: num -126.4 -103.8 -108.3 -102.7 -52.8 ...
## $ timbre_10_max: num 18.7 121.9 33.3 46.4 22.9 ...
## $ timbre_11_min: num -44.8 -38.9 -43.7 -59.4 -50.4 ...
## $ timbre_11_max: num 26 22.5 25.7 37.1 32.8 ...
## $ Top10 : int 0 0 0 0 0 0 0 0 0 1 ...
## NULL## Warning in myprint_str_df(df): [list output truncated]# glb_trnobs_df <- read.delim("data/hygiene.txt", header=TRUE, fill=TRUE, sep="\t",
# fileEncoding='iso-8859-1')
# glb_trnobs_df <- read.table("data/hygiene.dat.labels", col.names=c("dirty"),
# na.strings="[none]")
# glb_trnobs_df$review <- readLines("data/hygiene.dat", n =-1)
# comment(glb_trnobs_df) <- "glb_trnobs_df"
# glb_trnobs_df <- data.frame()
# for (symbol in c("Boeing", "CocaCola", "GE", "IBM", "ProcterGamble")) {
# sym_trnobs_df <-
# myimport_data(url=gsub("IBM", symbol, glb_trnng_url), comment="glb_trnobs_df",
# force_header=TRUE)
# sym_trnobs_df$Symbol <- symbol
# glb_trnobs_df <- myrbind_df(glb_trnobs_df, sym_trnobs_df)
# }
# glb_trnobs_df <-
# glb_trnobs_df %>% dplyr::filter(Year >= 1999)
if (glb_is_separate_newobs_dataset) {
glb_newobs_df <- myimport_data(url=glb_newdt_url, comment="glb_newobs_df",
force_header=TRUE)
# To make plots / stats / checks easier in chunk:inspectORexplore.data
glb_allobs_df <- myrbind_df(glb_trnobs_df, glb_newobs_df);
comment(glb_allobs_df) <- "glb_allobs_df"
} else {
glb_allobs_df <- glb_trnobs_df; comment(glb_allobs_df) <- "glb_allobs_df"
if (!glb_split_entity_newobs_datasets) {
stop("Not implemented yet")
glb_newobs_df <- glb_trnobs_df[sample(1:nrow(glb_trnobs_df),
max(2, nrow(glb_trnobs_df) / 1000)),]
} else if (glb_split_newdata_method == "condition") {
glb_newobs_df <- do.call("subset",
list(glb_trnobs_df, parse(text=glb_split_newdata_condition)))
glb_trnobs_df <- do.call("subset",
list(glb_trnobs_df, parse(text=paste0("!(",
glb_split_newdata_condition,
")"))))
} else if (glb_split_newdata_method == "sample") {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnobs_df[, glb_rsp_var_raw],
SplitRatio=(1-glb_split_newdata_size_ratio))
glb_newobs_df <- glb_trnobs_df[!split, ]
glb_trnobs_df <- glb_trnobs_df[split ,]
} else if (glb_split_newdata_method == "copy") {
glb_trnobs_df <- glb_allobs_df
comment(glb_trnobs_df) <- "glb_trnobs_df"
glb_newobs_df <- glb_allobs_df
comment(glb_newobs_df) <- "glb_newobs_df"
} else stop("glb_split_newdata_method should be %in% c('condition', 'sample', 'copy')")
comment(glb_newobs_df) <- "glb_newobs_df"
myprint_df(glb_newobs_df)
str(glb_newobs_df)
if (glb_split_entity_newobs_datasets) {
myprint_df(glb_trnobs_df)
str(glb_trnobs_df)
}
} ## year songtitle artistname
## 1 2010 This Is the House That Doubt Built A Day to Remember
## 2 2010 Sticks & Bricks A Day to Remember
## 3 2010 All I Want A Day to Remember
## 4 2010 It's Complicated A Day to Remember
## 5 2010 2nd Sucks A Day to Remember
## 6 2010 Better Off This Way A Day to Remember
## songID artistID timesignature
## 1 SOBGGAB12C5664F054 AROBSHL1187B9AFB01 3
## 2 SOPAQHU1315CD47F31 AROBSHL1187B9AFB01 4
## 3 SOOIZOU1376E7C6386 AROBSHL1187B9AFB01 4
## 4 SODRYWD1315CD49DBE AROBSHL1187B9AFB01 4
## 5 SOICMQB1315CD46EE3 AROBSHL1187B9AFB01 4
## 6 SOCEYON1315CD4A23E AROBSHL1187B9AFB01 4
## timesignature_confidence loudness tempo tempo_confidence key
## 1 0.853 -4.262 91.525 0.953 11
## 2 1.000 -4.051 140.048 0.921 10
## 3 1.000 -3.571 160.512 0.489 2
## 4 1.000 -3.815 97.525 0.794 1
## 5 0.788 -4.707 140.053 0.286 6
## 6 1.000 -3.807 160.366 0.347 4
## key_confidence energy pitch timbre_0_min timbre_0_max timbre_1_min
## 1 0.453 0.9666556 0.024 0.002 57.342 -6.496
## 2 0.469 0.9847095 0.025 0.000 57.414 -37.351
## 3 0.209 0.9899004 0.026 0.003 57.422 -17.222
## 4 0.632 0.9392072 0.013 0.000 57.765 -32.083
## 5 0.483 0.9877376 0.063 0.000 56.872 -223.922
## 6 0.627 0.9799530 0.038 0.000 57.083 -40.408
## timbre_1_max timbre_2_min timbre_2_max timbre_3_min timbre_3_max
## 1 171.093 -81.664 95.117 -285.049 259.426
## 2 171.130 -149.589 180.334 -380.143 384.166
## 3 171.060 -72.912 157.925 -203.984 251.258
## 4 220.895 -138.596 173.365 -73.490 373.492
## 5 171.130 -147.153 166.008 -128.082 389.419
## 6 174.002 -83.829 126.879 -100.119 173.607
## timbre_4_min timbre_4_max timbre_5_min timbre_5_max timbre_6_min
## 1 -40.385 73.630 -104.683 183.089 -88.771
## 2 -48.662 100.414 -87.267 42.798 -86.895
## 3 -66.044 152.095 -98.673 141.365 -88.874
## 4 -55.607 119.158 -77.515 141.178 -70.790
## 5 -43.908 99.332 -96.147 38.303 -110.757
## 6 -33.789 66.904 -84.451 47.268 -71.219
## timbre_6_max timbre_7_min timbre_7_max timbre_8_min timbre_8_max
## 1 73.549 -71.127 82.475 -52.025 39.116
## 2 75.455 -65.807 106.918 -61.320 35.378
## 3 66.504 -67.433 80.621 -59.773 45.979
## 4 64.540 -63.667 96.675 -78.660 41.088
## 5 72.391 -55.935 110.332 -56.450 37.555
## 6 71.239 -79.948 91.117 -54.378 53.808
## timbre_9_min timbre_9_max timbre_10_min timbre_10_max timbre_11_min
## 1 -35.368 71.642 -126.440 18.658 -44.770
## 2 -81.928 74.574 -103.808 121.935 -38.892
## 3 -46.293 59.904 -108.313 33.300 -43.733
## 4 -49.194 95.440 -102.676 46.422 -59.439
## 5 -48.588 67.570 -52.796 22.888 -50.414
## 6 -33.183 54.657 -64.478 34.522 -40.922
## timbre_11_max Top10
## 1 25.989 0
## 2 22.513 0
## 3 25.744 0
## 4 37.082 0
## 5 32.758 0
## 6 36.453 0
## year songtitle artistname
## 1 2010 This Is the House That Doubt Built A Day to Remember
## 66 2010 King Zero Drowning Pool
## 144 2010 Do-Wah-Doo Kate Nash
## 146 2010 I've Got A Secret Kate Nash
## 149 2010 Later On Kate Nash
## 172 2010 Bad Romance Lady Gaga
## songID artistID timesignature
## 1 SOBGGAB12C5664F054 AROBSHL1187B9AFB01 3
## 66 SORDZHA1375842B93A ARVSQHB1187B992FF0 4
## 144 SOBSGPF12AB018A2EC ARUZIPI1187FB5454A 4
## 146 SOOTJPM1315CD4E37E ARUZIPI1187FB5454A 4
## 149 SORAMKD131634AA473 ARUZIPI1187FB5454A 4
## 172 SOWPBHM13740123DE4 ARX6TAQ11C8A415850 4
## timesignature_confidence loudness tempo tempo_confidence key
## 1 0.853 -4.262 91.525 0.953 11
## 66 0.507 -4.482 170.038 0.347 4
## 144 1.000 -4.140 75.187 0.096 5
## 146 0.838 -7.300 134.372 0.300 7
## 149 0.928 -5.855 142.996 0.402 0
## 172 0.766 -5.280 121.123 0.757 9
## key_confidence energy pitch timbre_0_min timbre_0_max timbre_1_min
## 1 0.453 0.9666556 0.024 0.002 57.342 -6.496
## 66 0.008 0.9789121 0.034 0.000 56.858 -32.765
## 144 0.276 0.8982172 0.040 3.548 57.736 -65.857
## 146 0.174 0.6976624 0.008 18.959 55.941 -123.354
## 149 1.000 0.8296503 0.034 4.724 56.585 -110.203
## 172 0.206 0.9646445 0.010 28.933 57.919 -52.930
## timbre_1_max timbre_2_min timbre_2_max timbre_3_min timbre_3_max
## 1 171.093 -81.664 95.117 -285.049 259.426
## 66 236.966 -114.972 145.832 -291.344 249.504
## 144 209.740 -54.648 148.757 -110.410 235.860
## 146 150.717 -181.426 130.265 -144.576 351.420
## 149 143.525 -101.047 102.746 -215.517 126.843
## 172 188.946 -121.424 194.654 -319.608 124.892
## timbre_4_min timbre_4_max timbre_5_min timbre_5_max timbre_6_min
## 1 -40.385 73.630 -104.683 183.089 -88.771
## 66 -64.450 61.272 -89.033 166.842 -54.385
## 144 -25.883 125.275 -71.218 220.054 -78.952
## 146 -51.180 124.562 -119.357 207.993 -106.233
## 149 -44.579 142.076 -75.550 266.427 -56.031
## 172 -88.386 119.030 -112.415 217.453 -82.697
## timbre_6_max timbre_7_min timbre_7_max timbre_8_min timbre_8_max
## 1 73.549 -71.127 82.475 -52.025 39.116
## 66 60.711 -53.279 54.733 -52.828 38.089
## 144 68.543 -64.882 114.825 -75.785 23.124
## 146 86.776 -94.747 117.902 -70.564 38.605
## 149 62.192 -90.309 64.590 -88.586 41.525
## 172 101.543 -99.651 89.983 -89.644 41.225
## timbre_9_min timbre_9_max timbre_10_min timbre_10_max timbre_11_min
## 1 -35.368 71.642 -126.440 18.658 -44.770
## 66 -33.359 80.422 -202.256 135.774 -48.136
## 144 -63.403 64.000 -57.716 80.260 -33.235
## 146 -76.984 58.775 -131.919 49.218 -35.680
## 149 -50.205 80.019 -66.911 38.834 -50.193
## 172 -54.354 76.439 -86.599 48.143 -47.715
## timbre_11_max Top10
## 1 25.989 0
## 66 46.067 0
## 144 33.499 0
## 146 54.745 0
## 149 36.077 0
## 172 56.982 1
## year songtitle artistname songID
## 368 2010 Mighty to Save Various Artists SOZHGUW137375147FE
## 369 2010 All Because of Jesus Various Artists SOQNGML13DC8B9A8A5
## 370 2010 Your Name Various Artists SOXGDEA1375A1AD2F8
## 371 2010 You Reign Various Artists SOEHNSK13DC4A18344
## 372 2010 New Song We Sing Various Artists SOQOCRO13DC469AD34
## 373 2010 BedRock Young Money SOPQHEH1374011CFB8
## artistID timesignature timesignature_confidence loudness
## 368 ARAGWS81187FB3F768 4 0.465 -11.768
## 369 ARAGWS81187FB3F768 4 0.851 -6.896
## 370 ARAGWS81187FB3F768 4 0.937 -7.695
## 371 ARAGWS81187FB3F768 4 0.581 -13.761
## 372 ARAGWS81187FB3F768 4 0.869 -12.884
## 373 ARFNQBX122BCFCC40F 4 0.230 -5.860
## tempo tempo_confidence key key_confidence energy pitch
## 368 147.962 0.455 6 0.715 0.7076861 0.004
## 369 132.014 0.897 5 0.593 0.8218041 0.002
## 370 79.990 0.345 3 0.512 0.5038695 0.003
## 371 75.019 0.171 2 0.689 0.3286537 0.002
## 372 120.074 0.710 9 0.664 0.3213989 0.002
## 373 148.328 0.266 8 0.458 0.7864918 0.005
## timbre_0_min timbre_0_max timbre_1_min timbre_1_max timbre_2_min
## 368 13.320 51.565 -112.753 190.046 -109.409
## 369 0.000 55.832 -253.356 171.130 -113.093
## 370 0.008 57.291 -137.997 182.508 -108.564
## 371 0.000 52.977 -120.113 179.066 -99.847
## 372 0.000 50.539 -160.382 186.927 -135.012
## 373 33.782 56.072 -40.255 309.536 -131.923
## timbre_2_max timbre_3_min timbre_3_max timbre_4_min timbre_4_max
## 368 118.560 -144.309 125.791 -54.146 132.050
## 369 108.598 -152.949 94.648 -50.066 142.013
## 370 100.557 -230.832 74.961 -40.792 94.952
## 371 143.332 -209.714 131.918 -37.509 132.303
## 372 196.199 -182.354 229.050 -33.858 128.405
## 373 126.088 -159.948 169.983 -54.751 148.642
## timbre_5_min timbre_5_max timbre_6_min timbre_6_max timbre_7_min
## 368 -88.464 77.963 -69.467 69.589 -58.550
## 369 -80.187 42.780 -53.975 70.978 -113.356
## 370 -95.304 157.062 -84.841 73.270 -92.767
## 371 -86.539 176.159 -85.282 96.528 -123.525
## 372 -87.455 158.989 -135.819 69.098 -94.179
## 373 -111.185 125.831 -111.483 72.581 -73.553
## timbre_7_max timbre_8_min timbre_8_max timbre_9_min timbre_9_max
## 368 86.250 -72.253 61.809 -55.720 66.168
## 369 122.253 -50.909 61.278 -65.765 61.128
## 370 91.915 -78.383 29.942 -62.093 47.771
## 371 100.910 -100.195 50.555 -64.052 60.195
## 372 99.038 -53.112 42.895 -64.875 99.467
## 373 88.488 -70.974 51.550 -51.106 85.348
## timbre_10_min timbre_10_max timbre_11_min timbre_11_max Top10
## 368 -63.432 44.960 -61.604 48.130 0
## 369 -54.461 32.120 -46.549 32.486 0
## 370 -49.670 51.155 -44.724 41.997 0
## 371 -83.234 51.323 -59.022 63.993 0
## 372 -111.697 49.754 -39.289 69.057 0
## 373 -84.959 74.872 -57.014 51.651 1
## 'data.frame': 373 obs. of 39 variables:
## $ year : int 2010 2010 2010 2010 2010 2010 2010 2010 2010 2010 ...
## $ songtitle : chr "This Is the House That Doubt Built" "Sticks & Bricks" "All I Want" "It's Complicated" ...
## $ artistname : chr "A Day to Remember" "A Day to Remember" "A Day to Remember" "A Day to Remember" ...
## $ songID : chr "SOBGGAB12C5664F054" "SOPAQHU1315CD47F31" "SOOIZOU1376E7C6386" "SODRYWD1315CD49DBE" ...
## $ artistID : chr "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" "AROBSHL1187B9AFB01" ...
## $ timesignature : int 3 4 4 4 4 4 4 4 4 4 ...
## $ timesignature_confidence: num 0.853 1 1 1 0.788 1 0.968 0.861 0.622 0.938 ...
## $ loudness : num -4.26 -4.05 -3.57 -3.81 -4.71 ...
## $ tempo : num 91.5 140 160.5 97.5 140.1 ...
## $ tempo_confidence : num 0.953 0.921 0.489 0.794 0.286 0.347 0.273 0.83 0.018 0.929 ...
## $ key : int 11 10 2 1 6 4 10 5 9 11 ...
## $ key_confidence : num 0.453 0.469 0.209 0.632 0.483 0.627 0.715 0.423 0.751 0.602 ...
## $ energy : num 0.967 0.985 0.99 0.939 0.988 ...
## $ pitch : num 0.024 0.025 0.026 0.013 0.063 0.038 0.026 0.033 0.027 0.004 ...
## $ timbre_0_min : num 0.002 0 0.003 0 0 ...
## $ timbre_0_max : num 57.3 57.4 57.4 57.8 56.9 ...
## $ timbre_1_min : num -6.5 -37.4 -17.2 -32.1 -223.9 ...
## $ timbre_1_max : num 171 171 171 221 171 ...
## $ timbre_2_min : num -81.7 -149.6 -72.9 -138.6 -147.2 ...
## $ timbre_2_max : num 95.1 180.3 157.9 173.4 166 ...
## $ timbre_3_min : num -285 -380.1 -204 -73.5 -128.1 ...
## $ timbre_3_max : num 259 384 251 373 389 ...
## $ timbre_4_min : num -40.4 -48.7 -66 -55.6 -43.9 ...
## $ timbre_4_max : num 73.6 100.4 152.1 119.2 99.3 ...
## $ timbre_5_min : num -104.7 -87.3 -98.7 -77.5 -96.1 ...
## $ timbre_5_max : num 183.1 42.8 141.4 141.2 38.3 ...
## $ timbre_6_min : num -88.8 -86.9 -88.9 -70.8 -110.8 ...
## $ timbre_6_max : num 73.5 75.5 66.5 64.5 72.4 ...
## $ timbre_7_min : num -71.1 -65.8 -67.4 -63.7 -55.9 ...
## $ timbre_7_max : num 82.5 106.9 80.6 96.7 110.3 ...
## $ timbre_8_min : num -52 -61.3 -59.8 -78.7 -56.5 ...
## $ timbre_8_max : num 39.1 35.4 46 41.1 37.6 ...
## $ timbre_9_min : num -35.4 -81.9 -46.3 -49.2 -48.6 ...
## $ timbre_9_max : num 71.6 74.6 59.9 95.4 67.6 ...
## $ timbre_10_min : num -126.4 -103.8 -108.3 -102.7 -52.8 ...
## $ timbre_10_max : num 18.7 121.9 33.3 46.4 22.9 ...
## $ timbre_11_min : num -44.8 -38.9 -43.7 -59.4 -50.4 ...
## $ timbre_11_max : num 26 22.5 25.7 37.1 32.8 ...
## $ Top10 : int 0 0 0 0 0 0 0 0 0 1 ...
## - attr(*, "comment")= chr "glb_newobs_df"
## year songtitle artistname songID
## 374 2009 The Awkward Goodbye Athlete SOUALGK12AB017FC37
## 375 2009 Rubik's Cube Athlete SOGPIQC12AB0182B15
## 376 2009 Superhuman Touch Athlete SOBNYZN13774E81F76
## 377 2009 The Getaway Athlete SOHFEOA1366EE931DD
## 378 2009 Black Swan Song Athlete SOXXSMX12AB017F7B3
## 379 2009 Don't Hold Your Breath Athlete SOOEDWA12AB017FC13
## artistID timesignature timesignature_confidence loudness
## 374 ARDW3YJ1187FB4CCE5 3 0.732 -6.320
## 375 ARDW3YJ1187FB4CCE5 3 0.906 -9.541
## 376 ARDW3YJ1187FB4CCE5 4 0.987 -4.842
## 377 ARDW3YJ1187FB4CCE5 4 0.822 -5.272
## 378 ARDW3YJ1187FB4CCE5 4 0.983 -6.233
## 379 ARDW3YJ1187FB4CCE5 4 1.000 -6.793
## tempo tempo_confidence key key_confidence energy pitch
## 374 89.614 0.652 1 0.773 0.5985294 0.004
## 375 117.742 0.542 0 0.722 0.3633990 0.006
## 376 119.018 0.838 6 0.106 0.7601515 0.003
## 377 71.479 0.613 4 0.781 0.7550336 0.014
## 378 77.492 0.740 8 0.552 0.5236583 0.008
## 379 81.859 0.821 9 0.218 0.6546091 0.012
## timbre_0_min timbre_0_max timbre_1_min timbre_1_max timbre_2_min
## 374 0.000 57.831 -62.306 285.818 -81.802
## 375 0.739 57.059 -220.205 241.091 -96.833
## 376 0.000 57.815 -189.660 187.282 -139.053
## 377 0.000 58.330 -113.885 171.130 -71.640
## 378 0.000 57.643 -160.579 216.778 -79.456
## 379 0.000 57.389 -103.691 227.209 -155.016
## timbre_2_max timbre_3_min timbre_3_max timbre_4_min timbre_4_max
## 374 211.084 -217.025 203.151 -55.874 97.646
## 375 214.510 -201.889 124.200 -52.389 131.859
## 376 134.508 -116.316 94.698 -55.617 79.292
## 377 194.788 -276.297 146.268 -59.374 121.707
## 378 114.093 -183.559 108.719 -31.922 169.734
## 379 174.758 -386.464 185.756 -69.700 103.106
## timbre_5_min timbre_5_max timbre_6_min timbre_6_max timbre_7_min
## 374 -62.492 82.169 -82.129 59.197 -109.384
## 375 -73.875 73.628 -63.496 70.133 -90.092
## 376 -73.474 41.025 -41.489 62.759 -69.311
## 377 -71.135 39.607 -77.786 94.525 -69.088
## 378 -73.004 233.930 -76.026 58.016 -78.803
## 379 -75.267 184.309 -62.755 45.283 -61.878
## timbre_7_max timbre_8_min timbre_8_max timbre_9_min timbre_9_max
## 374 70.975 -71.776 58.432 -53.816 88.571
## 375 112.879 -64.470 58.086 -76.937 74.441
## 376 90.400 -52.459 40.679 -50.408 58.811
## 377 93.373 -55.811 78.963 -51.504 70.455
## 378 100.766 -61.392 50.309 -62.994 96.837
## 379 89.443 -70.718 50.515 -54.980 80.278
## timbre_10_min timbre_10_max timbre_11_min timbre_11_max Top10
## 374 -89.816 38.026 -52.075 52.827 0
## 375 -88.244 42.209 -66.812 40.749 0
## 376 -78.239 35.264 -54.200 46.490 0
## 377 -74.928 30.839 -51.377 27.768 0
## 378 -90.397 60.549 -52.122 48.059 0
## 379 -70.317 54.084 -43.651 43.261 0
## year songtitle artistname
## 2726 2005 American Guns Transplants
## 3640 2003 Don't Let the World Get in Your Way The Jayhawks
## 5012 1999 Even Odds Jeff Beck
## 5465 1998 Phone Call Skit Queen Latifah
## 7226 1991 Ellen West Throwing Muses
## 7323 1990 Epic Faith No More
## songID artistID timesignature
## 2726 SOASWBQ13134394F22 ARF4WYR1187FB3838D 4
## 3640 SORBFQV1313438E57F AR4TV7J1187FB400C9 4
## 5012 SOFNLJE12AB01801EB ARHI20L1187FB4C997 4
## 5465 SOFLVRM1315CD46AA2 ARELPXQ1187FB384FD 3
## 7226 SOQJGUA1315CD40B11 ARWV4ZQ1187FB4A44F 4
## 7323 SODICGU1373F1D2CE3 AR0AJG11187FB51EFC 4
## timesignature_confidence loudness tempo tempo_confidence key
## 2726 1.000 -5.635 93.369 1.000 5
## 3640 0.947 -5.965 83.460 0.792 4
## 5012 0.364 -7.764 91.902 0.975 0
## 5465 0.750 -16.109 90.736 0.854 1
## 7226 0.988 -11.877 121.219 0.824 4
## 7323 1.000 -7.368 87.187 0.914 4
## key_confidence energy pitch timbre_0_min timbre_0_max timbre_1_min
## 2726 0.991 0.9742504 0.038 1.530 56.356 -33.893
## 3640 0.400 0.3074298 0.005 0.991 58.787 -208.433
## 5012 0.370 0.8056250 0.014 0.951 55.084 1.988
## 5465 0.254 0.3204597 0.021 8.616 45.455 -270.967
## 7226 0.458 0.4965909 0.007 0.000 50.967 -108.305
## 7323 0.809 0.8953855 0.012 0.000 55.115 -130.600
## timbre_1_max timbre_2_min timbre_2_max timbre_3_min timbre_3_max
## 2726 174.678 -61.115 174.919 -129.347 117.431
## 3640 185.035 -159.470 88.993 -175.936 92.497
## 5012 217.262 -144.212 101.045 -254.606 285.427
## 5465 198.236 -162.788 121.035 -125.385 172.790
## 7226 171.130 -175.175 98.137 -261.209 287.941
## 7323 230.373 -183.707 110.840 -239.676 143.577
## timbre_4_min timbre_4_max timbre_5_min timbre_5_max timbre_6_min
## 2726 -30.230 88.339 -73.830 144.070 -62.843
## 3640 -76.347 131.807 -72.179 97.136 -97.176
## 5012 -47.381 178.111 -96.249 191.226 -121.316
## 5465 -90.194 130.643 -78.386 75.061 -74.612
## 7226 -68.878 95.331 -111.167 186.491 -86.676
## 7323 -64.810 131.815 -117.699 162.767 -88.636
## timbre_6_max timbre_7_min timbre_7_max timbre_8_min timbre_8_max
## 2726 39.098 -62.324 71.844 -71.110 44.070
## 3640 51.147 -77.978 86.337 -45.057 52.735
## 5012 61.408 -111.579 99.271 -37.422 58.007
## 5465 93.092 -66.750 88.664 -76.198 37.251
## 7226 64.456 -60.156 116.897 -75.238 47.249
## 7323 71.788 -64.151 146.595 -79.778 49.692
## timbre_9_min timbre_9_max timbre_10_min timbre_10_max timbre_11_min
## 2726 -33.131 49.908 -95.315 33.230 -29.731
## 3640 -55.272 52.952 -66.075 31.141 -58.252
## 5012 -28.135 82.095 -113.360 119.897 -52.073
## 5465 -45.053 84.435 -75.624 65.293 -22.330
## 7226 -42.790 37.726 -113.553 62.800 -44.657
## 7323 -50.031 88.352 -122.997 81.541 -49.090
## timbre_11_max Top10
## 2726 27.310 0
## 3640 45.012 0
## 5012 38.193 0
## 5465 43.326 0
## 7226 57.052 0
## 7323 74.035 1
## year songtitle artistname songID
## 7569 1990 Red Hot & Ready Y&T SOHVWXR12A6D4FC59C
## 7570 1990 She's Gone Y&T SOSIEQB12A6D4FC59D
## 7571 1990 Let It Out Y&T SOIGIQI12A6D4FC59E
## 7572 1990 Ten Lovers Y&T SOLNEQO12A6D4FC59F
## 7573 1990 Goin' Off The Deep End Y&T SOEMJEP12A58A7E7B5
## 7574 1990 Surrender Y&T SOLBQSX12A6D4FC5A0
## artistID timesignature timesignature_confidence loudness
## 7569 ARGQANQ11F50C4769E 4 1.000 -10.574
## 7570 ARGQANQ11F50C4769E 4 1.000 -10.197
## 7571 ARGQANQ11F50C4769E 4 1.000 -12.392
## 7572 ARGQANQ11F50C4769E 4 0.984 -10.304
## 7573 ARGQANQ11F50C4769E 4 0.907 -9.295
## 7574 ARGQANQ11F50C4769E 4 0.987 -9.762
## tempo tempo_confidence key key_confidence energy pitch
## 7569 137.872 1.000 2 0.006 0.9736458 0.030
## 7570 93.140 0.859 5 0.889 0.9429917 0.016
## 7571 79.858 0.196 9 0.149 0.8124216 0.012
## 7572 91.760 0.592 2 0.077 0.7368709 0.016
## 7573 110.907 0.838 9 0.621 0.9900532 0.061
## 7574 139.650 0.781 6 0.193 0.9448227 0.027
## timbre_0_min timbre_0_max timbre_1_min timbre_1_max timbre_2_min
## 7569 0 51.847 -18.801 203.289 -141.762
## 7570 0 53.462 -22.878 202.424 -77.467
## 7571 0 51.354 -91.916 202.639 -76.736
## 7572 0 53.358 -10.087 202.877 -48.768
## 7573 0 52.928 -15.289 175.845 -119.044
## 7574 0 52.989 -32.907 195.601 -146.973
## timbre_2_max timbre_3_min timbre_3_max timbre_4_min timbre_4_max
## 7569 157.204 -120.035 364.391 -53.240 99.468
## 7570 174.663 -170.802 201.677 -45.807 109.383
## 7571 165.643 -186.653 166.200 -63.367 132.495
## 7572 194.665 -201.008 132.174 -55.916 129.674
## 7573 164.155 -162.944 391.622 -42.728 137.159
## 7574 150.596 -79.356 307.651 -39.215 68.426
## timbre_5_min timbre_5_max timbre_6_min timbre_6_max timbre_7_min
## 7569 -82.554 115.220 -91.407 62.922 -57.069
## 7570 -75.011 103.305 -130.215 59.775 -61.197
## 7571 -115.231 86.509 -83.905 102.373 -66.416
## 7572 -82.989 166.003 -83.246 62.951 -69.512
## 7573 -80.171 92.551 -64.419 74.428 -38.794
## 7574 -77.927 202.200 -55.617 98.615 -71.984
## timbre_7_max timbre_8_min timbre_8_max timbre_9_min timbre_9_max
## 7569 96.283 -56.468 34.205 -50.403 51.369
## 7570 38.120 -66.638 50.105 -40.692 43.763
## 7571 83.454 -83.707 44.124 -51.628 69.929
## 7572 103.413 -101.464 36.152 -45.387 48.352
## 7573 108.688 -55.893 42.222 -76.631 68.336
## 7574 87.098 -48.440 52.198 -60.673 48.418
## timbre_10_min timbre_10_max timbre_11_min timbre_11_max Top10
## 7569 -63.965 82.665 -47.429 58.158 0
## 7570 -59.707 49.414 -53.970 68.303 0
## 7571 -97.153 36.745 -61.243 56.902 0
## 7572 -57.103 67.641 -53.729 65.176 0
## 7573 -83.284 56.476 -51.687 59.427 0
## 7574 -120.625 49.593 -47.656 70.005 0
## 'data.frame': 7201 obs. of 39 variables:
## $ year : int 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 ...
## $ songtitle : chr "The Awkward Goodbye" "Rubik's Cube" "Superhuman Touch" "The Getaway" ...
## $ artistname : chr "Athlete" "Athlete" "Athlete" "Athlete" ...
## $ songID : chr "SOUALGK12AB017FC37" "SOGPIQC12AB0182B15" "SOBNYZN13774E81F76" "SOHFEOA1366EE931DD" ...
## $ artistID : chr "ARDW3YJ1187FB4CCE5" "ARDW3YJ1187FB4CCE5" "ARDW3YJ1187FB4CCE5" "ARDW3YJ1187FB4CCE5" ...
## $ timesignature : int 3 3 4 4 4 4 4 4 4 4 ...
## $ timesignature_confidence: num 0.732 0.906 0.987 0.822 0.983 1 0.821 0.997 0.816 1 ...
## $ loudness : num -6.32 -9.54 -4.84 -5.27 -6.23 ...
## $ tempo : num 89.6 117.7 119 71.5 77.5 ...
## $ tempo_confidence : num 0.652 0.542 0.838 0.613 0.74 0.821 0.912 0.609 0.786 0.27 ...
## $ key : int 1 0 6 4 8 9 6 9 0 9 ...
## $ key_confidence : num 0.773 0.722 0.106 0.781 0.552 0.218 0.275 0.333 0.634 0.578 ...
## $ energy : num 0.599 0.363 0.76 0.755 0.524 ...
## $ pitch : num 0.004 0.006 0.003 0.014 0.008 0.012 0.002 0.003 0.001 0.006 ...
## $ timbre_0_min : num 0 0.739 0 0 0 ...
## $ timbre_0_max : num 57.8 57.1 57.8 58.3 57.6 ...
## $ timbre_1_min : num -62.3 -220.2 -189.7 -113.9 -160.6 ...
## $ timbre_1_max : num 286 241 187 171 217 ...
## $ timbre_2_min : num -81.8 -96.8 -139.1 -71.6 -79.5 ...
## $ timbre_2_max : num 211 215 135 195 114 ...
## $ timbre_3_min : num -217 -202 -116 -276 -184 ...
## $ timbre_3_max : num 203.2 124.2 94.7 146.3 108.7 ...
## $ timbre_4_min : num -55.9 -52.4 -55.6 -59.4 -31.9 ...
## $ timbre_4_max : num 97.6 131.9 79.3 121.7 169.7 ...
## $ timbre_5_min : num -62.5 -73.9 -73.5 -71.1 -73 ...
## $ timbre_5_max : num 82.2 73.6 41 39.6 233.9 ...
## $ timbre_6_min : num -82.1 -63.5 -41.5 -77.8 -76 ...
## $ timbre_6_max : num 59.2 70.1 62.8 94.5 58 ...
## $ timbre_7_min : num -109.4 -90.1 -69.3 -69.1 -78.8 ...
## $ timbre_7_max : num 71 112.9 90.4 93.4 100.8 ...
## $ timbre_8_min : num -71.8 -64.5 -52.5 -55.8 -61.4 ...
## $ timbre_8_max : num 58.4 58.1 40.7 79 50.3 ...
## $ timbre_9_min : num -53.8 -76.9 -50.4 -51.5 -63 ...
## $ timbre_9_max : num 88.6 74.4 58.8 70.5 96.8 ...
## $ timbre_10_min : num -89.8 -88.2 -78.2 -74.9 -90.4 ...
## $ timbre_10_max : num 38 42.2 35.3 30.8 60.5 ...
## $ timbre_11_min : num -52.1 -66.8 -54.2 -51.4 -52.1 ...
## $ timbre_11_max : num 52.8 40.7 46.5 27.8 48.1 ...
## $ Top10 : int 0 0 0 0 0 0 0 0 0 0 ...if ((num_nas <- sum(is.na(glb_trnobs_df[, glb_rsp_var_raw]))) > 0)
stop("glb_trnobs_df$", glb_rsp_var_raw, " contains NAs for ", num_nas, " obs")
if (nrow(glb_trnobs_df) == nrow(glb_allobs_df))
warning("glb_trnobs_df same as glb_allobs_df")
if (nrow(glb_newobs_df) == nrow(glb_allobs_df))
warning("glb_newobs_df same as glb_allobs_df")
if (length(glb_drop_vars) > 0) {
warning("dropping vars: ", paste0(glb_drop_vars, collapse=", "))
glb_allobs_df <- glb_allobs_df[, setdiff(names(glb_allobs_df), glb_drop_vars)]
glb_trnobs_df <- glb_trnobs_df[, setdiff(names(glb_trnobs_df), glb_drop_vars)]
glb_newobs_df <- glb_newobs_df[, setdiff(names(glb_newobs_df), glb_drop_vars)]
}
#stop(here"); sav_allobs_df <- glb_allobs_df # glb_allobs_df <- sav_allobs_df
# Combine trnent & newobs into glb_allobs_df for easier manipulation
glb_trnobs_df$.src <- "Train"; glb_newobs_df$.src <- "Test";
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, ".src")
glb_allobs_df <- myrbind_df(glb_trnobs_df, glb_newobs_df)
comment(glb_allobs_df) <- "glb_allobs_df"
# Check for duplicates in glb_id_var
if (length(glb_id_var) == 0) {
warning("using .rownames as identifiers for observations")
glb_allobs_df$.rownames <- rownames(glb_allobs_df)
glb_trnobs_df$.rownames <- rownames(subset(glb_allobs_df, .src == "Train"))
glb_newobs_df$.rownames <- rownames(subset(glb_allobs_df, .src == "Test"))
glb_id_var <- ".rownames"
}## Warning: using .rownames as identifiers for observationsif (sum(duplicated(glb_allobs_df[, glb_id_var, FALSE])) > 0)
stop(glb_id_var, " duplicated in glb_allobs_df")
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_id_var)
glb_allobs_df <- orderBy(reformulate(glb_id_var), glb_allobs_df)
glb_trnobs_df <- glb_newobs_df <- NULL
glb_chunks_df <- myadd_chunk(glb_chunks_df, "inspect.data", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 1 import.data 1 0 8.268 9.832 1.564
## 2 inspect.data 2 0 9.832 NA NAStep 2.0: inspect data
#print(str(glb_allobs_df))
#View(glb_allobs_df)
dsp_class_dstrb <- function(var) {
xtab_df <- mycreate_xtab_df(glb_allobs_df, c(".src", var))
rownames(xtab_df) <- xtab_df$.src
xtab_df <- subset(xtab_df, select=-.src)
print(xtab_df)
print(xtab_df / rowSums(xtab_df, na.rm=TRUE))
}
# Performed repeatedly in other chunks
glb_chk_data <- function() {
# Histogram of predictor in glb_trnobs_df & glb_newobs_df
print(myplot_histogram(glb_allobs_df, glb_rsp_var_raw) + facet_wrap(~ .src))
if (glb_is_classification)
dsp_class_dstrb(var=ifelse(glb_rsp_var %in% names(glb_allobs_df),
glb_rsp_var, glb_rsp_var_raw))
mycheck_problem_data(glb_allobs_df)
}
glb_chk_data()## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## stat_bin: binwidth defaulted to range/30. Use 'binwidth = x' to adjust this.
## Loading required package: reshape2
## Top10.0 Top10.1
## Test 314 59
## Train 6141 1060
## Top10.0 Top10.1
## Test 0.8418231 0.1581769
## Train 0.8527982 0.1472018
## [1] "numeric data missing in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ 0s in glb_allobs_df: "
## timesignature timesignature_confidence tempo
## 10 157 13
## tempo_confidence key key_confidence
## 85 803 422
## pitch timbre_0_min Top10
## 108 3114 6455
## [1] "numeric data w/ Infs in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ NaNs in glb_allobs_df: "
## named integer(0)
## [1] "string data missing in glb_allobs_df: "
## songtitle artistname songID artistID .rownames
## 0 0 0 0 0# Create new features that help diagnostics
if (!is.null(glb_map_rsp_raw_to_var)) {
glb_allobs_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_allobs_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_allobs_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
if (glb_is_classification) dsp_class_dstrb(glb_rsp_var)
}## Loading required package: sqldf
## Loading required package: gsubfn
## Loading required package: proto
## Loading required package: RSQLite
## Loading required package: DBI
## Loading required package: tcltk## Top10 Top10.fctr .n
## 1 0 N 6455
## 2 1 Y 1119
## Top10.fctr.N Top10.fctr.Y
## Test 314 59
## Train 6141 1060
## Top10.fctr.N Top10.fctr.Y
## Test 0.8418231 0.1581769
## Train 0.8527982 0.1472018# check distribution of all numeric data
dsp_numeric_feats_dstrb <- function(feats_vctr) {
for (feat in feats_vctr) {
print(sprintf("feat: %s", feat))
if (glb_is_regression)
gp <- myplot_scatter(df=glb_allobs_df, ycol_name=glb_rsp_var, xcol_name=feat,
smooth=TRUE)
if (glb_is_classification)
gp <- myplot_box(df=glb_allobs_df, ycol_names=feat, xcol_name=glb_rsp_var)
if (inherits(glb_allobs_df[, feat], "factor"))
gp <- gp + facet_wrap(reformulate(feat))
print(gp)
}
}
# dsp_numeric_vars_dstrb(setdiff(names(glb_allobs_df),
# union(myfind_chr_cols_df(glb_allobs_df),
# c(glb_rsp_var_raw, glb_rsp_var))))
add_new_diag_feats <- function(obs_df, ref_df=glb_allobs_df) {
require(plyr)
obs_df <- mutate(obs_df,
# <col_name>.NA=is.na(<col_name>),
# <col_name>.fctr=factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# <col_name>.fctr=relevel(factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# "<ref_val>"),
# <col2_name>.fctr=relevel(factor(ifelse(<col1_name> == <val>, "<oth_val>", "<ref_val>")),
# as.factor(c("R", "<ref_val>")),
# ref="<ref_val>"),
# This doesn't work - use sapply instead
# <col_name>.fctr_num=grep(<col_name>, levels(<col_name>.fctr)),
#
# Date.my=as.Date(strptime(Date, "%m/%d/%y %H:%M")),
# Year=year(Date.my),
# Month=months(Date.my),
# Weekday=weekdays(Date.my)
# <col_name>=<table>[as.character(<col2_name>)],
# <col_name>=as.numeric(<col2_name>),
# <col_name> = trunc(<col2_name> / 100),
.rnorm = rnorm(n=nrow(obs_df))
)
# If levels of a factor are different across obs_df & glb_newobs_df; predict.glm fails
# Transformations not handled by mutate
# obs_df$<col_name>.fctr.num <- sapply(1:nrow(obs_df),
# function(row_ix) grep(obs_df[row_ix, "<col_name>"],
# levels(obs_df[row_ix, "<col_name>.fctr"])))
#print(summary(obs_df))
#print(sapply(names(obs_df), function(col) sum(is.na(obs_df[, col]))))
return(obs_df)
}
glb_allobs_df <- add_new_diag_feats(glb_allobs_df)## Loading required package: plyrrequire(dplyr)## Loading required package: dplyr
##
## Attaching package: 'dplyr'
##
## The following objects are masked from 'package:plyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize
##
## The following objects are masked from 'package:stats':
##
## filter, lag
##
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union#stop(here"); sav_allobs_df <- glb_allobs_df # glb_allobs_df <- sav_allobs_df
# Merge some <descriptor>
# glb_allobs_df$<descriptor>.my <- glb_allobs_df$<descriptor>
# glb_allobs_df[grepl("\\bAIRPORT\\b", glb_allobs_df$<descriptor>.my),
# "<descriptor>.my"] <- "AIRPORT"
# glb_allobs_df$<descriptor>.my <-
# plyr::revalue(glb_allobs_df$<descriptor>.my, c(
# "ABANDONED BUILDING" = "OTHER",
# "##" = "##"
# ))
# print(<descriptor>_freq_df <- mycreate_sqlxtab_df(glb_allobs_df, c("<descriptor>.my")))
# # print(dplyr::filter(<descriptor>_freq_df, grepl("(MEDICAL|DENTAL|OFFICE)", <descriptor>.my)))
# # print(dplyr::filter(dplyr::select(glb_allobs_df, -<var.zoo>),
# # grepl("STORE", <descriptor>.my)))
# glb_exclude_vars_as_features <- c(glb_exclude_vars_as_features, "<descriptor>")
# Check distributions of newly transformed / extracted vars
# Enhancement: remove vars that were displayed ealier
dsp_numeric_feats_dstrb(feats_vctr=setdiff(names(glb_allobs_df),
c(myfind_chr_cols_df(glb_allobs_df), glb_rsp_var_raw, glb_rsp_var,
glb_exclude_vars_as_features)))## [1] "feat: timesignature"
## [1] "feat: timesignature_confidence"
## [1] "feat: loudness"
## [1] "feat: tempo"
## [1] "feat: tempo_confidence"
## [1] "feat: key"
## [1] "feat: key_confidence"
## [1] "feat: energy"
## [1] "feat: pitch"
## [1] "feat: timbre_0_min"
## [1] "feat: timbre_0_max"
## [1] "feat: timbre_1_min"
## [1] "feat: timbre_1_max"
## [1] "feat: timbre_2_min"
## [1] "feat: timbre_2_max"
## [1] "feat: timbre_3_min"
## [1] "feat: timbre_3_max"
## [1] "feat: timbre_4_min"
## [1] "feat: timbre_4_max"
## [1] "feat: timbre_5_min"
## [1] "feat: timbre_5_max"
## [1] "feat: timbre_6_min"
## [1] "feat: timbre_6_max"
## [1] "feat: timbre_7_min"
## [1] "feat: timbre_7_max"
## [1] "feat: timbre_8_min"
## [1] "feat: timbre_8_max"
## [1] "feat: timbre_9_min"
## [1] "feat: timbre_9_max"
## [1] "feat: timbre_10_min"
## [1] "feat: timbre_10_max"
## [1] "feat: timbre_11_min"
## [1] "feat: timbre_11_max"
## [1] "feat: .rnorm"
# Convert factors to dummy variables
# Build splines require(splines); bsBasis <- bs(training$age, df=3)
#pairs(subset(glb_trnobs_df, select=-c(col_symbol)))
# Check for glb_newobs_df & glb_trnobs_df features range mismatches
# Other diagnostics:
# print(subset(glb_trnobs_df, <col1_name> == max(glb_trnobs_df$<col1_name>, na.rm=TRUE) &
# <col2_name> <= mean(glb_trnobs_df$<col1_name>, na.rm=TRUE)))
# print(glb_trnobs_df[which.max(glb_trnobs_df$<col_name>),])
# print(<col_name>_freq_glb_trnobs_df <- mycreate_tbl_df(glb_trnobs_df, "<col_name>"))
# print(which.min(table(glb_trnobs_df$<col_name>)))
# print(which.max(table(glb_trnobs_df$<col_name>)))
# print(which.max(table(glb_trnobs_df$<col1_name>, glb_trnobs_df$<col2_name>)[, 2]))
# print(table(glb_trnobs_df$<col1_name>, glb_trnobs_df$<col2_name>))
# print(table(is.na(glb_trnobs_df$<col1_name>), glb_trnobs_df$<col2_name>))
# print(table(sign(glb_trnobs_df$<col1_name>), glb_trnobs_df$<col2_name>))
# print(mycreate_xtab_df(glb_trnobs_df, <col1_name>))
# print(mycreate_xtab_df(glb_trnobs_df, c(<col1_name>, <col2_name>)))
# print(<col1_name>_<col2_name>_xtab_glb_trnobs_df <-
# mycreate_xtab_df(glb_trnobs_df, c("<col1_name>", "<col2_name>")))
# <col1_name>_<col2_name>_xtab_glb_trnobs_df[is.na(<col1_name>_<col2_name>_xtab_glb_trnobs_df)] <- 0
# print(<col1_name>_<col2_name>_xtab_glb_trnobs_df <-
# mutate(<col1_name>_<col2_name>_xtab_glb_trnobs_df,
# <col3_name>=(<col1_name> * 1.0) / (<col1_name> + <col2_name>)))
# print(mycreate_sqlxtab_df(glb_allobs_df, c("<col1_name>", "<col2_name>")))
# print(<col2_name>_min_entity_arr <-
# sort(tapply(glb_trnobs_df$<col1_name>, glb_trnobs_df$<col2_name>, min, na.rm=TRUE)))
# print(<col1_name>_na_by_<col2_name>_arr <-
# sort(tapply(glb_trnobs_df$<col1_name>.NA, glb_trnobs_df$<col2_name>, mean, na.rm=TRUE)))
# Other plots:
# print(myplot_box(df=glb_trnobs_df, ycol_names="<col1_name>"))
# print(myplot_box(df=glb_trnobs_df, ycol_names="<col1_name>", xcol_name="<col2_name>"))
# print(myplot_line(subset(glb_trnobs_df, Symbol %in% c("CocaCola", "ProcterGamble")),
# "Date.POSIX", "StockPrice", facet_row_colnames="Symbol") +
# geom_vline(xintercept=as.numeric(as.POSIXlt("2003-03-01"))) +
# geom_vline(xintercept=as.numeric(as.POSIXlt("1983-01-01")))
# )
# print(myplot_line(subset(glb_trnobs_df, Date.POSIX > as.POSIXct("2004-01-01")),
# "Date.POSIX", "StockPrice") +
# geom_line(aes(color=Symbol)) +
# coord_cartesian(xlim=c(as.POSIXct("1990-01-01"),
# as.POSIXct("2000-01-01"))) +
# coord_cartesian(ylim=c(0, 250)) +
# geom_vline(xintercept=as.numeric(as.POSIXlt("1997-09-01"))) +
# geom_vline(xintercept=as.numeric(as.POSIXlt("1997-11-01")))
# )
# print(myplot_scatter(glb_allobs_df, "<col1_name>", "<col2_name>", smooth=TRUE))
# print(myplot_scatter(glb_allobs_df, "<col1_name>", "<col2_name>", colorcol_name="<Pred.fctr>") +
# geom_point(data=subset(glb_allobs_df, <condition>),
# mapping=aes(x=<x_var>, y=<y_var>), color="red", shape=4, size=5) +
# geom_vline(xintercept=84))
glb_chunks_df <- myadd_chunk(glb_chunks_df, "scrub.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 2 inspect.data 2 0 9.832 27.612 17.78
## 3 scrub.data 2 1 27.612 NA NAStep 2.1: scrub data
mycheck_problem_data(glb_allobs_df)## [1] "numeric data missing in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ 0s in glb_allobs_df: "
## timesignature timesignature_confidence tempo
## 10 157 13
## tempo_confidence key key_confidence
## 85 803 422
## pitch timbre_0_min Top10
## 108 3114 6455
## [1] "numeric data w/ Infs in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ NaNs in glb_allobs_df: "
## named integer(0)
## [1] "string data missing in glb_allobs_df: "
## songtitle artistname songID artistID .rownames
## 0 0 0 0 0dsp_catgs <- function() {
print("NewsDesk:")
print(table(glb_allobs_df$NewsDesk))
print("SectionName:")
print(table(glb_allobs_df$SectionName))
print("SubsectionName:")
print(table(glb_allobs_df$SubsectionName))
}
# sel_obs <- function(Popular=NULL,
# NewsDesk=NULL, SectionName=NULL, SubsectionName=NULL,
# Headline.contains=NULL, Snippet.contains=NULL, Abstract.contains=NULL,
# Headline.pfx=NULL, NewsDesk.nb=NULL, .clusterid=NULL, myCategory=NULL,
# perl=FALSE) {
sel_obs <- function(vars_lst) {
tmp_df <- glb_allobs_df
# Does not work for Popular == NAs ???
if (!is.null(Popular)) {
if (is.na(Popular))
tmp_df <- tmp_df[is.na(tmp_df$Popular), ] else
tmp_df <- tmp_df[tmp_df$Popular == Popular, ]
}
if (!is.null(NewsDesk))
tmp_df <- tmp_df[tmp_df$NewsDesk == NewsDesk, ]
if (!is.null(SectionName))
tmp_df <- tmp_df[tmp_df$SectionName == SectionName, ]
if (!is.null(SubsectionName))
tmp_df <- tmp_df[tmp_df$SubsectionName == SubsectionName, ]
if (!is.null(Headline.contains))
tmp_df <-
tmp_df[grep(Headline.contains, tmp_df$Headline, perl=perl), ]
if (!is.null(Snippet.contains))
tmp_df <-
tmp_df[grep(Snippet.contains, tmp_df$Snippet, perl=perl), ]
if (!is.null(Abstract.contains))
tmp_df <-
tmp_df[grep(Abstract.contains, tmp_df$Abstract, perl=perl), ]
if (!is.null(Headline.pfx)) {
if (length(grep("Headline.pfx", names(tmp_df), fixed=TRUE, value=TRUE))
> 0) tmp_df <-
tmp_df[tmp_df$Headline.pfx == Headline.pfx, ] else
warning("glb_allobs_df does not contain Headline.pfx; ignoring that filter")
}
if (!is.null(NewsDesk.nb)) {
if (any(grepl("NewsDesk.nb", names(tmp_df), fixed=TRUE)) > 0)
tmp_df <-
tmp_df[tmp_df$NewsDesk.nb == NewsDesk.nb, ] else
warning("glb_allobs_df does not contain NewsDesk.nb; ignoring that filter")
}
if (!is.null(.clusterid)) {
if (any(grepl(".clusterid", names(tmp_df), fixed=TRUE)) > 0)
tmp_df <-
tmp_df[tmp_df$clusterid == clusterid, ] else
warning("glb_allobs_df does not contain clusterid; ignoring that filter") }
if (!is.null(myCategory)) {
if (!(myCategory %in% names(glb_allobs_df)))
tmp_df <-
tmp_df[tmp_df$myCategory == myCategory, ] else
warning("glb_allobs_df does not contain myCategory; ignoring that filter")
}
return(glb_allobs_df$UniqueID %in% tmp_df$UniqueID)
}
dsp_obs <- function(..., cols=c(NULL), all=FALSE) {
tmp_df <- glb_allobs_df[sel_obs(...),
union(c("UniqueID", "Popular", "myCategory", "Headline"), cols), FALSE]
if(all) { print(tmp_df) } else { myprint_df(tmp_df) }
}
#dsp_obs(Popular=1, NewsDesk="", SectionName="", Headline.contains="Boehner")
# dsp_obs(Popular=1, NewsDesk="", SectionName="")
# dsp_obs(Popular=NA, NewsDesk="", SectionName="")
dsp_tbl <- function(...) {
tmp_entity_df <- glb_allobs_df[sel_obs(...), ]
tmp_tbl <- table(tmp_entity_df$NewsDesk,
tmp_entity_df$SectionName,
tmp_entity_df$SubsectionName,
tmp_entity_df$Popular, useNA="ifany")
#print(names(tmp_tbl))
#print(dimnames(tmp_tbl))
print(tmp_tbl)
}
dsp_hdlxtab <- function(str)
print(mycreate_sqlxtab_df(glb_allobs_df[sel_obs(Headline.contains=str), ],
c("Headline.pfx", "Headline", glb_rsp_var)))
#dsp_hdlxtab("(1914)|(1939)")
dsp_catxtab <- function(str)
print(mycreate_sqlxtab_df(glb_allobs_df[sel_obs(Headline.contains=str), ],
c("Headline.pfx", "NewsDesk", "SectionName", "SubsectionName", glb_rsp_var)))
# dsp_catxtab("1914)|(1939)")
# dsp_catxtab("19(14|39|64):")
# dsp_catxtab("19..:")
# Create myCategory <- NewsDesk#SectionName#SubsectionName
# Fix some data before merging categories
# glb_allobs_df[sel_obs(Headline.contains="Your Turn:", NewsDesk=""),
# "NewsDesk"] <- "Styles"
# glb_allobs_df[sel_obs(Headline.contains="School", NewsDesk="", SectionName="U.S.",
# SubsectionName=""),
# "SubsectionName"] <- "Education"
# glb_allobs_df[sel_obs(Headline.contains="Today in Small Business:", NewsDesk="Business"),
# "SectionName"] <- "Business Day"
# glb_allobs_df[sel_obs(Headline.contains="Today in Small Business:", NewsDesk="Business"),
# "SubsectionName"] <- "Small Business"
# glb_allobs_df[sel_obs(Headline.contains="Readers Respond:"),
# "SectionName"] <- "Opinion"
# glb_allobs_df[sel_obs(Headline.contains="Readers Respond:"),
# "SubsectionName"] <- "Room For Debate"
# glb_allobs_df[sel_obs(NewsDesk="Business", SectionName="", SubsectionName="", Popular=NA),
# "SubsectionName"] <- "Small Business"
# print(glb_allobs_df[glb_allobs_df$UniqueID %in% c(7973),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
# glb_allobs_df[sel_obs(NewsDesk="Business", SectionName="", SubsectionName=""),
# "SectionName"] <- "Technology"
# print(glb_allobs_df[glb_allobs_df$UniqueID %in% c(5076, 5736, 5924, 5911, 6532),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
# glb_allobs_df[sel_obs(SectionName="Health"),
# "NewsDesk"] <- "Science"
# glb_allobs_df[sel_obs(SectionName="Travel"),
# "NewsDesk"] <- "Travel"
#
# glb_allobs_df[sel_obs(SubsectionName="Fashion & Style"),
# "SectionName"] <- ""
# glb_allobs_df[sel_obs(SubsectionName="Fashion & Style"),
# "SubsectionName"] <- ""
# glb_allobs_df[sel_obs(NewsDesk="Styles", SectionName="", SubsectionName="", Popular=1),
# "SectionName"] <- "U.S."
# print(glb_allobs_df[glb_allobs_df$UniqueID %in% c(5486),
# c("UniqueID", "Headline", "myCategory", "NewsDesk", "SectionName", "SubsectionName")])
#
# glb_allobs_df$myCategory <- paste(glb_allobs_df$NewsDesk,
# glb_allobs_df$SectionName,
# glb_allobs_df$SubsectionName,
# sep="#")
# dsp_obs( Headline.contains="Music:"
# #,NewsDesk=""
# #,SectionName=""
# #,SubsectionName="Fashion & Style"
# #,Popular=1 #NA
# ,cols= c("UniqueID", "Headline", "Popular", "myCategory",
# "NewsDesk", "SectionName", "SubsectionName"),
# all=TRUE)
# dsp_obs( Headline.contains="."
# ,NewsDesk=""
# ,SectionName="Opinion"
# ,SubsectionName=""
# #,Popular=1 #NA
# ,cols= c("UniqueID", "Headline", "Popular", "myCategory",
# "NewsDesk", "SectionName", "SubsectionName"),
# all=TRUE)
# Merge some categories
# glb_allobs_df$myCategory <-
# plyr::revalue(glb_allobs_df$myCategory, c(
# "#Business Day#Dealbook" = "Business#Business Day#Dealbook",
# "#Business Day#Small Business" = "Business#Business Day#Small Business",
# "#Crosswords/Games#" = "Business#Crosswords/Games#",
# "Business##" = "Business#Technology#",
# "#Open#" = "Business#Technology#",
# "#Technology#" = "Business#Technology#",
#
# "#Arts#" = "Culture#Arts#",
# "Culture##" = "Culture#Arts#",
#
# "#World#Asia Pacific" = "Foreign#World#Asia Pacific",
# "Foreign##" = "Foreign#World#",
#
# "#N.Y. / Region#" = "Metro#N.Y. / Region#",
#
# "#Opinion#" = "OpEd#Opinion#",
# "OpEd##" = "OpEd#Opinion#",
#
# "#Health#" = "Science#Health#",
# "Science##" = "Science#Health#",
#
# "Styles##" = "Styles##Fashion",
# "Styles#Health#" = "Science#Health#",
# "Styles#Style#Fashion & Style" = "Styles##Fashion",
#
# "#Travel#" = "Travel#Travel#",
#
# "Magazine#Magazine#" = "myOther",
# "National##" = "myOther",
# "National#U.S.#Politics" = "myOther",
# "Sports##" = "myOther",
# "Sports#Sports#" = "myOther",
# "#U.S.#" = "myOther",
#
#
# # "Business##Small Business" = "Business#Business Day#Small Business",
# #
# # "#Opinion#" = "#Opinion#Room For Debate",
# "##" = "##"
# # "Business##" = "Business#Business Day#Dealbook",
# # "Foreign#World#" = "Foreign##",
# # "#Open#" = "Other",
# # "#Opinion#The Public Editor" = "OpEd#Opinion#",
# # "Styles#Health#" = "Styles##",
# # "Styles#Style#Fashion & Style" = "Styles##",
# # "#U.S.#" = "#U.S.#Education",
# ))
# ctgry_xtab_df <- orderBy(reformulate(c("-", ".n")),
# mycreate_sqlxtab_df(glb_allobs_df,
# c("myCategory", "NewsDesk", "SectionName", "SubsectionName", glb_rsp_var)))
# myprint_df(ctgry_xtab_df)
# write.table(ctgry_xtab_df, paste0(glb_out_pfx, "ctgry_xtab.csv"),
# row.names=FALSE)
# ctgry_cast_df <- orderBy(~ -Y -NA, dcast(ctgry_xtab_df,
# myCategory + NewsDesk + SectionName + SubsectionName ~
# Popular.fctr, sum, value.var=".n"))
# myprint_df(ctgry_cast_df)
# write.table(ctgry_cast_df, paste0(glb_out_pfx, "ctgry_cast.csv"),
# row.names=FALSE)
# print(ctgry_sum_tbl <- table(glb_allobs_df$myCategory, glb_allobs_df[, glb_rsp_var],
# useNA="ifany"))
dsp_chisq.test <- function(...) {
sel_df <- glb_allobs_df[sel_obs(...) &
!is.na(glb_allobs_df$Popular), ]
sel_df$.marker <- 1
ref_df <- glb_allobs_df[!is.na(glb_allobs_df$Popular), ]
mrg_df <- merge(ref_df[, c(glb_id_var, "Popular")],
sel_df[, c(glb_id_var, ".marker")], all.x=TRUE)
mrg_df[is.na(mrg_df)] <- 0
print(mrg_tbl <- table(mrg_df$.marker, mrg_df$Popular))
print("Rows:Selected; Cols:Popular")
#print(mrg_tbl)
print(chisq.test(mrg_tbl))
}
# dsp_chisq.test(Headline.contains="[Ee]bola")
# dsp_chisq.test(Snippet.contains="[Ee]bola")
# dsp_chisq.test(Abstract.contains="[Ee]bola")
# print(mycreate_sqlxtab_df(glb_allobs_df[sel_obs(Headline.contains="[Ee]bola"), ],
# c(glb_rsp_var, "NewsDesk", "SectionName", "SubsectionName")))
# print(table(glb_allobs_df$NewsDesk, glb_allobs_df$SectionName))
# print(table(glb_allobs_df$SectionName, glb_allobs_df$SubsectionName))
# print(table(glb_allobs_df$NewsDesk, glb_allobs_df$SectionName, glb_allobs_df$SubsectionName))
# glb_allobs_df$myCategory.fctr <- as.factor(glb_allobs_df$myCategory)
# glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
# c("myCategory", "NewsDesk", "SectionName", "SubsectionName"))
# Copy Headline into Snipper & Abstract if they are empty
# print(glb_allobs_df[nchar(glb_allobs_df[, "Snippet"]) == 0, c("Headline", "Snippet")])
# print(glb_allobs_df[glb_allobs_df$Headline == glb_allobs_df$Snippet,
# c("UniqueID", "Headline", "Snippet")])
# glb_allobs_df[nchar(glb_allobs_df[, "Snippet"]) == 0, "Snippet"] <-
# glb_allobs_df[nchar(glb_allobs_df[, "Snippet"]) == 0, "Headline"]
#
# print(glb_allobs_df[nchar(glb_allobs_df[, "Abstract"]) == 0, c("Headline", "Abstract")])
# print(glb_allobs_df[glb_allobs_df$Headline == glb_allobs_df$Abstract,
# c("UniqueID", "Headline", "Abstract")])
# glb_allobs_df[nchar(glb_allobs_df[, "Abstract"]) == 0, "Abstract"] <-
# glb_allobs_df[nchar(glb_allobs_df[, "Abstract"]) == 0, "Headline"]
# WordCount_0_df <- subset(glb_allobs_df, WordCount == 0)
# table(WordCount_0_df$Popular, WordCount_0_df$WordCount, useNA="ifany")
# myprint_df(WordCount_0_df[,
# c("UniqueID", "Popular", "WordCount", "Headline")])Step 2.1: scrub data
glb_chunks_df <- myadd_chunk(glb_chunks_df, "transform.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 3 scrub.data 2 1 27.612 31.328 3.716
## 4 transform.data 2 2 31.328 NA NA### Mapping dictionary
#sav_allobs_df <- glb_allobs_df; glb_allobs_df <- sav_allobs_df
if (!is.null(glb_map_vars)) {
for (feat in glb_map_vars) {
map_df <- myimport_data(url=glb_map_urls[[feat]],
comment="map_df",
print_diagn=TRUE)
glb_allobs_df <- mymap_codes(glb_allobs_df, feat, names(map_df)[2],
map_df, map_join_col_name=names(map_df)[1],
map_tgt_col_name=names(map_df)[2])
}
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_map_vars)
}
### Forced Assignments
#stop(here"); sav_allobs_df <- glb_allobs_df; glb_allobs_df <- sav_allobs_df
for (feat in glb_assign_vars) {
new_feat <- paste0(feat, ".my")
print(sprintf("Forced Assignments for: %s -> %s...", feat, new_feat))
glb_allobs_df[, new_feat] <- glb_allobs_df[, feat]
pairs <- glb_assign_pairs_lst[[feat]]
for (pair_ix in 1:length(pairs$from)) {
if (is.na(pairs$from[pair_ix]))
nobs <- nrow(filter(glb_allobs_df,
is.na(eval(parse(text=feat),
envir=glb_allobs_df)))) else
nobs <- sum(glb_allobs_df[, feat] == pairs$from[pair_ix])
#nobs <- nrow(filter(glb_allobs_df, is.na(Married.fctr))) ; print(nobs)
if ((is.na(pairs$from[pair_ix])) && (is.na(pairs$to[pair_ix])))
stop("what are you trying to do ???")
if (is.na(pairs$from[pair_ix]))
glb_allobs_df[is.na(glb_allobs_df[, feat]), new_feat] <-
pairs$to[pair_ix] else
glb_allobs_df[glb_allobs_df[, feat] == pairs$from[pair_ix], new_feat] <-
pairs$to[pair_ix]
print(sprintf(" %s -> %s for %s obs",
pairs$from[pair_ix], pairs$to[pair_ix], format(nobs, big.mark=",")))
}
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_assign_vars)
}
### Derivations using mapping functions
#stop(here"); sav_allobs_df <- glb_allobs_df; glb_allobs_df <- sav_allobs_df
for (new_feat in glb_derive_vars) {
print(sprintf("Creating new feature: %s...", new_feat))
args_lst <- NULL
for (arg in glb_derive_lst[[new_feat]]$args)
args_lst[[arg]] <- glb_allobs_df[, arg]
glb_allobs_df[, new_feat] <- do.call(glb_derive_lst[[new_feat]]$mapfn, args_lst)
}Step 2.2: transform data
#```{r extract_features, cache=FALSE, eval=!is.null(glb_txt_vars)}
glb_chunks_df <- myadd_chunk(glb_chunks_df, "extract.features", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 4 transform.data 2 2 31.328 31.388 0.06
## 5 extract.features 3 0 31.389 NA NAextract.features_chunk_df <- myadd_chunk(NULL, "extract.features_bgn")## label step_major step_minor bgn end elapsed
## 1 extract.features_bgn 1 0 31.395 NA NA# Options:
# Select Tf, log(1 + Tf), Tf-IDF or BM25Tf-IDf
# Create new features that help prediction
# <col_name>.lag.2 <- lag(zoo(glb_trnobs_df$<col_name>), -2, na.pad=TRUE)
# glb_trnobs_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
# <col_name>.lag.2 <- lag(zoo(glb_newobs_df$<col_name>), -2, na.pad=TRUE)
# glb_newobs_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
#
# glb_newobs_df[1, "<col_name>.lag.2"] <- glb_trnobs_df[nrow(glb_trnobs_df) - 1,
# "<col_name>"]
# glb_newobs_df[2, "<col_name>.lag.2"] <- glb_trnobs_df[nrow(glb_trnobs_df),
# "<col_name>"]
# glb_allobs_df <- mutate(glb_allobs_df,
# A.P.http=ifelse(grepl("http",Added,fixed=TRUE), 1, 0)
# )
#
# glb_trnobs_df <- mutate(glb_trnobs_df,
# )
#
# glb_newobs_df <- mutate(glb_newobs_df,
# )
# Convert dates to numbers
# typically, dates come in as chars;
# so this must be done before converting chars to factors
#stop(here"); sav_allobs_df <- glb_allobs_df #; glb_allobs_df <- sav_allobs_df
if (!is.null(glb_date_vars)) {
glb_allobs_df <- cbind(glb_allobs_df,
myextract_dates_df(df=glb_allobs_df, vars=glb_date_vars,
id_vars=glb_id_var, rsp_var=glb_rsp_var))
for (sfx in c("", ".POSIX"))
glb_exclude_vars_as_features <-
union(glb_exclude_vars_as_features,
paste(glb_date_vars, sfx, sep=""))
for (feat in glb_date_vars) {
glb_allobs_df <- orderBy(reformulate(paste0(feat, ".POSIX")), glb_allobs_df)
# print(myplot_scatter(glb_allobs_df, xcol_name=paste0(feat, ".POSIX"),
# ycol_name=glb_rsp_var, colorcol_name=glb_rsp_var))
print(myplot_scatter(glb_allobs_df[glb_allobs_df[, paste0(feat, ".POSIX")] >=
strptime("2012-12-01", "%Y-%m-%d"), ],
xcol_name=paste0(feat, ".POSIX"),
ycol_name=glb_rsp_var, colorcol_name=paste0(feat, ".wkend")))
# Create features that measure the gap between previous timestamp in the data
require(zoo)
z <- zoo(as.numeric(as.POSIXlt(glb_allobs_df[, paste0(feat, ".POSIX")])))
glb_allobs_df[, paste0(feat, ".zoo")] <- z
print(head(glb_allobs_df[, c(glb_id_var, feat, paste0(feat, ".zoo"))]))
print(myplot_scatter(glb_allobs_df[glb_allobs_df[, paste0(feat, ".POSIX")] >
strptime("2012-10-01", "%Y-%m-%d"), ],
xcol_name=paste0(feat, ".zoo"), ycol_name=glb_rsp_var,
colorcol_name=glb_rsp_var))
b <- zoo(, seq(nrow(glb_allobs_df)))
last1 <- as.numeric(merge(z-lag(z, -1), b, all=TRUE)); last1[is.na(last1)] <- 0
glb_allobs_df[, paste0(feat, ".last1.log")] <- log(1 + last1)
print(gp <- myplot_box(df=glb_allobs_df[glb_allobs_df[,
paste0(feat, ".last1.log")] > 0, ],
ycol_names=paste0(feat, ".last1.log"),
xcol_name=glb_rsp_var))
last2 <- as.numeric(merge(z-lag(z, -2), b, all=TRUE)); last2[is.na(last2)] <- 0
glb_allobs_df[, paste0(feat, ".last2.log")] <- log(1 + last2)
print(gp <- myplot_box(df=glb_allobs_df[glb_allobs_df[,
paste0(feat, ".last2.log")] > 0, ],
ycol_names=paste0(feat, ".last2.log"),
xcol_name=glb_rsp_var))
last10 <- as.numeric(merge(z-lag(z, -10), b, all=TRUE)); last10[is.na(last10)] <- 0
glb_allobs_df[, paste0(feat, ".last10.log")] <- log(1 + last10)
print(gp <- myplot_box(df=glb_allobs_df[glb_allobs_df[,
paste0(feat, ".last10.log")] > 0, ],
ycol_names=paste0(feat, ".last10.log"),
xcol_name=glb_rsp_var))
last100 <- as.numeric(merge(z-lag(z, -100), b, all=TRUE)); last100[is.na(last100)] <- 0
glb_allobs_df[, paste0(feat, ".last100.log")] <- log(1 + last100)
print(gp <- myplot_box(df=glb_allobs_df[glb_allobs_df[,
paste0(feat, ".last100.log")] > 0, ],
ycol_names=paste0(feat, ".last100.log"),
xcol_name=glb_rsp_var))
glb_allobs_df <- orderBy(reformulate(glb_id_var), glb_allobs_df)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c(paste0(feat, ".zoo")))
# all2$last3 = as.numeric(merge(z-lag(z, -3), b, all = TRUE))
# all2$last5 = as.numeric(merge(z-lag(z, -5), b, all = TRUE))
# all2$last10 = as.numeric(merge(z-lag(z, -10), b, all = TRUE))
# all2$last20 = as.numeric(merge(z-lag(z, -20), b, all = TRUE))
# all2$last50 = as.numeric(merge(z-lag(z, -50), b, all = TRUE))
#
#
# # order table
# all2 = all2[order(all2$id),]
#
# ## fill in NAs
# # count averages
# na.avg = all2 %>% group_by(weekend, hour) %>% dplyr::summarise(
# last1=mean(last1, na.rm=TRUE),
# last3=mean(last3, na.rm=TRUE),
# last5=mean(last5, na.rm=TRUE),
# last10=mean(last10, na.rm=TRUE),
# last20=mean(last20, na.rm=TRUE),
# last50=mean(last50, na.rm=TRUE)
# )
#
# # fill in averages
# na.merge = merge(all2, na.avg, by=c("weekend","hour"))
# na.merge = na.merge[order(na.merge$id),]
# for(i in c("last1", "last3", "last5", "last10", "last20", "last50")) {
# y = paste0(i, ".y")
# idx = is.na(all2[[i]])
# all2[idx,][[i]] <- na.merge[idx,][[y]]
# }
# rm(na.avg, na.merge, b, i, idx, n, pd, sec, sh, y, z)
}
}
rm(last1, last10, last100)## Warning in rm(last1, last10, last100): object 'last1' not found## Warning in rm(last1, last10, last100): object 'last10' not found## Warning in rm(last1, last10, last100): object 'last100' not found# Create factors of string variables
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "factorize.str.vars"), major.inc=TRUE)## label step_major step_minor bgn end
## 1 extract.features_bgn 1 0 31.395 31.409
## 2 extract.features_factorize.str.vars 2 0 31.409 NA
## elapsed
## 1 0.014
## 2 NA#stop(here"); sav_allobs_df <- glb_allobs_df; #glb_allobs_df <- sav_allobs_df
print(str_vars <- myfind_chr_cols_df(glb_allobs_df))## songtitle artistname songID artistID .src
## "songtitle" "artistname" "songID" "artistID" ".src"
## .rownames
## ".rownames"if (length(str_vars <- setdiff(str_vars,
c(glb_exclude_vars_as_features, glb_txt_vars))) > 0) {
for (var in str_vars) {
warning("Creating factors of string variable: ", var,
": # of unique values: ", length(unique(glb_allobs_df[, var])))
glb_allobs_df[, paste0(var, ".fctr")] <-
relevel(factor(glb_allobs_df[, var]),
names(which.max(table(glb_allobs_df[, var], useNA = "ifany"))))
}
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, str_vars)
}
if (!is.null(glb_txt_vars)) {
require(foreach)
require(gsubfn)
require(stringr)
require(tm)
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "process.text"), major.inc=TRUE)
chk_pattern_freq <- function(rex_str, ignore.case=TRUE) {
match_mtrx <- str_extract_all(txt_vctr, regex(rex_str, ignore_case=ignore.case),
simplify=TRUE)
match_df <- as.data.frame(match_mtrx[match_mtrx != ""])
names(match_df) <- "pattern"
return(mycreate_sqlxtab_df(match_df, "pattern"))
}
# match_lst <- gregexpr("\\bok(?!ay)", txt_vctr[746], ignore.case = FALSE, perl=TRUE); print(match_lst)
dsp_pattern <- function(rex_str, ignore.case=TRUE, print.all=TRUE) {
match_lst <- gregexpr(rex_str, txt_vctr, ignore.case = ignore.case, perl=TRUE)
match_lst <- regmatches(txt_vctr, match_lst)
match_df <- data.frame(matches=sapply(match_lst,
function (elems) paste(elems, collapse="#")))
match_df <- subset(match_df, matches != "")
if (print.all)
print(match_df)
return(match_df)
}
dsp_matches <- function(rex_str, ix) {
print(match_pos <- gregexpr(rex_str, txt_vctr[ix], perl=TRUE))
print(str_sub(txt_vctr[ix], (match_pos[[1]] / 100) * 99 + 0,
(match_pos[[1]] / 100) * 100 + 100))
}
myapply_gsub <- function(...) {
if ((length_lst <- length(names(gsub_map_lst))) == 0)
return(txt_vctr)
for (ptn_ix in 1:length_lst) {
if ((ptn_ix %% 10) == 0)
print(sprintf("running gsub for %02d (of %02d): #%s#...", ptn_ix,
length(names(gsub_map_lst)), names(gsub_map_lst)[ptn_ix]))
txt_vctr <- gsub(names(gsub_map_lst)[ptn_ix], gsub_map_lst[[ptn_ix]],
txt_vctr, ...)
}
return(txt_vctr)
}
myapply_txtmap <- function(txt_vctr, ...) {
nrows <- nrow(glb_txt_map_df)
for (ptn_ix in 1:nrows) {
if ((ptn_ix %% 10) == 0)
print(sprintf("running gsub for %02d (of %02d): #%s#...", ptn_ix,
nrows, glb_txt_map_df[ptn_ix, "rex_str"]))
txt_vctr <- gsub(glb_txt_map_df[ptn_ix, "rex_str"],
glb_txt_map_df[ptn_ix, "rpl_str"],
txt_vctr, ...)
}
return(txt_vctr)
}
chk.equal <- function(bgn, end) {
print(all.equal(sav_txt_lst[["Headline"]][bgn:end],
glb_txt_lst[["Headline"]][bgn:end]))
}
dsp.equal <- function(bgn, end) {
print(sav_txt_lst[["Headline"]][bgn:end])
print(glb_txt_lst[["Headline"]][bgn:end])
}
#sav_txt_lst <- glb_txt_lst; all.equal(sav_txt_lst, glb_txt_lst)
#all.equal(sav_txt_lst[["Headline"]][1:4200], glb_txt_lst[["Headline"]][1:4200])
#chk.equal( 1, 100)
#dsp.equal(86, 90)
glb_txt_map_df <- read.csv("mytxt_map.csv", comment.char="#", strip.white=TRUE)
glb_txt_lst <- list();
print(sprintf("Building glb_txt_lst..."))
glb_txt_lst <- foreach(txt_var=glb_txt_vars) %dopar% {
# for (txt_var in glb_txt_vars) {
txt_vctr <- glb_allobs_df[, txt_var]
# myapply_txtmap shd be created as a tm_map::content_transformer ?
#print(glb_txt_map_df)
#txt_var=glb_txt_vars[3]; txt_vctr <- glb_txt_lst[[txt_var]]
#print(rex_str <- glb_txt_map_df[163, "rex_str"])
#print(rex_str <- glb_txt_map_df[glb_txt_map_df$rex_str == "\\bWall St\\.", "rex_str"])
#print(rex_str <- glb_txt_map_df[grepl("du Pont", glb_txt_map_df$rex_str), "rex_str"])
#print(rex_str <- glb_txt_map_df[glb_txt_map_df$rpl_str == "versus", "rex_str"])
#print(tmp_vctr <- grep(rex_str, txt_vctr, value=TRUE, ignore.case=FALSE))
#ret_lst <- regexec(rex_str, txt_vctr, ignore.case=FALSE); ret_lst <- regmatches(txt_vctr, ret_lst); ret_vctr <- sapply(1:length(ret_lst), function(pos_ix) ifelse(length(ret_lst[[pos_ix]]) > 0, ret_lst[[pos_ix]], "")); print(ret_vctr <- ret_vctr[ret_vctr != ""])
#gsub(rex_str, glb_txt_map_df[glb_txt_map_df$rex_str == rex_str, "rpl_str"], tmp_vctr, ignore.case=FALSE)
#grep("Hong Hong", txt_vctr, value=TRUE)
txt_vctr <- myapply_txtmap(txt_vctr, ignore.case=FALSE)
}
names(glb_txt_lst) <- glb_txt_vars
for (txt_var in glb_txt_vars) {
print(sprintf("Remaining OK in %s:", txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
print(chk_pattern_freq(rex_str <- "(?<!(BO|HO|LO))OK(?!(E\\!|ED|IE|IN|S ))",
ignore.case=FALSE))
match_df <- dsp_pattern(rex_str, ignore.case=FALSE, print.all=FALSE)
for (row in row.names(match_df))
dsp_matches(rex_str, ix=as.numeric(row))
print(chk_pattern_freq(rex_str <- "Ok(?!(a\\.|ay|in|ra|um))", ignore.case=FALSE))
match_df <- dsp_pattern(rex_str, ignore.case=FALSE, print.all=FALSE)
for (row in row.names(match_df))
dsp_matches(rex_str, ix=as.numeric(row))
print(chk_pattern_freq(rex_str <- "(?<!( b| B| c| C| g| G| j| M| p| P| w| W| r| Z|\\(b|ar|bo|Bo|co|Co|Ew|gk|go|ho|ig|jo|kb|ke|Ke|ki|lo|Lo|mo|mt|no|No|po|ra|ro|sm|Sm|Sp|to|To))ok(?!(ay|bo|e |e\\)|e,|e\\.|eb|ed|el|en|er|es|ey|i |ie|in|it|ka|ke|ki|ly|on|oy|ra|st|u |uc|uy|yl|yo))",
ignore.case=FALSE))
match_df <- dsp_pattern(rex_str, ignore.case=FALSE, print.all=FALSE)
for (row in row.names(match_df))
dsp_matches(rex_str, ix=as.numeric(row))
}
# txt_vctr <- glb_txt_lst[[glb_txt_vars[1]]]
# print(chk_pattern_freq(rex_str <- "(?<!( b| c| C| p|\\(b|bo|co|lo|Lo|Sp|to|To))ok(?!(ay|e |e\\)|e,|e\\.|ed|el|en|es|ey|ie|in|on|ra))", ignore.case=FALSE))
# print(chk_pattern_freq(rex_str <- "ok(?!(ay|el|on|ra))", ignore.case=FALSE))
# dsp_pattern(rex_str, ignore.case=FALSE, print.all=FALSE)
# dsp_matches(rex_str, ix=8)
# substr(txt_vctr[86], 5613, 5620)
# substr(glb_allobs_df[301, "review"], 550, 650)
#stop(here"); sav_txt_lst <- glb_txt_lst
for (txt_var in glb_txt_vars) {
print(sprintf("Remaining Acronyms in %s:", txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
print(chk_pattern_freq(rex_str <- "([[:upper:]]\\.( *)){2,}", ignore.case=FALSE))
# Check for names
print(subset(chk_pattern_freq(rex_str <- "(([[:upper:]]+)\\.( *)){1}",
ignore.case=FALSE),
.n > 1))
# dsp_pattern(rex_str="(OK\\.( *)){1}", ignore.case=FALSE)
# dsp_matches(rex_str="(OK\\.( *)){1}", ix=557)
#dsp_matches(rex_str="\\bR\\.I\\.P(\\.*)(\\B)", ix=461)
#dsp_matches(rex_str="\\bR\\.I\\.P(\\.*)", ix=461)
#print(str_sub(txt_vctr[676], 10100, 10200))
#print(str_sub(txt_vctr[74], 1, -1))
}
for (txt_var in glb_txt_vars) {
re_str <- "\\b(Fort|Ft\\.|Hong|Las|Los|New|Puerto|Saint|San|St\\.)( |-)(\\w)+"
print(sprintf("Remaining #%s# terms in %s: ", re_str, txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
print(orderBy(~ -.n +pattern, subset(chk_pattern_freq(re_str, ignore.case=FALSE),
grepl("( |-)[[:upper:]]", pattern))))
print(" consider cleaning if relevant to problem domain; geography name; .n > 1")
#grep("New G", txt_vctr, value=TRUE, ignore.case=FALSE)
#grep("St\\. Wins", txt_vctr, value=TRUE, ignore.case=FALSE)
}
#stop(here"); sav_txt_lst <- glb_txt_lst
for (txt_var in glb_txt_vars) {
re_str <- "\\b(N|S|E|W|C)( |\\.)(\\w)+"
print(sprintf("Remaining #%s# terms in %s: ", re_str, txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
print(orderBy(~ -.n +pattern, subset(chk_pattern_freq(re_str, ignore.case=FALSE),
grepl(".", pattern))))
#grep("N Weaver", txt_vctr, value=TRUE, ignore.case=FALSE)
}
for (txt_var in glb_txt_vars) {
re_str <- "\\b(North|South|East|West|Central)( |\\.)(\\w)+"
print(sprintf("Remaining #%s# terms in %s: ", re_str, txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
print(orderBy(~ -.n +pattern, subset(chk_pattern_freq(re_str, ignore.case=FALSE),
grepl(".", pattern))))
#grep("Central (African|Bankers|Cast|Italy|Role|Spring)", txt_vctr, value=TRUE, ignore.case=FALSE)
#grep("East (Africa|Berlin|London|Poland|Rivals|Spring)", txt_vctr, value=TRUE, ignore.case=FALSE)
#grep("North (American|Korean|West)", txt_vctr, value=TRUE, ignore.case=FALSE)
#grep("South (Pacific|Street)", txt_vctr, value=TRUE, ignore.case=FALSE)
#grep("St\\. Martins", txt_vctr, value=TRUE, ignore.case=FALSE)
}
find_cmpnd_wrds <- function(txt_vctr) {
txt_corpus <- Corpus(VectorSource(txt_vctr))
txt_corpus <- tm_map(txt_corpus, tolower)
txt_corpus <- tm_map(txt_corpus, PlainTextDocument)
txt_corpus <- tm_map(txt_corpus, removePunctuation,
preserve_intra_word_dashes=TRUE)
full_Tf_DTM <- DocumentTermMatrix(txt_corpus,
control=list(weighting=weightTf))
print(" Full TermMatrix:"); print(full_Tf_DTM)
full_Tf_mtrx <- as.matrix(full_Tf_DTM)
rownames(full_Tf_mtrx) <- rownames(glb_allobs_df) # print undreadable otherwise
full_Tf_vctr <- colSums(full_Tf_mtrx)
names(full_Tf_vctr) <- dimnames(full_Tf_DTM)[[2]]
#grep("year", names(full_Tf_vctr), value=TRUE)
#which.max(full_Tf_mtrx[, "yearlong"])
full_Tf_df <- as.data.frame(full_Tf_vctr)
names(full_Tf_df) <- "Tf.full"
full_Tf_df$term <- rownames(full_Tf_df)
#full_Tf_df$freq.full <- colSums(full_Tf_mtrx != 0)
full_Tf_df <- orderBy(~ -Tf.full, full_Tf_df)
cmpnd_Tf_df <- full_Tf_df[grep("-", full_Tf_df$term, value=TRUE) ,]
filter_df <- read.csv("mytxt_compound.csv", comment.char="#", strip.white=TRUE)
cmpnd_Tf_df$filter <- FALSE
for (row_ix in 1:nrow(filter_df))
cmpnd_Tf_df[!cmpnd_Tf_df$filter, "filter"] <-
grepl(filter_df[row_ix, "rex_str"],
cmpnd_Tf_df[!cmpnd_Tf_df$filter, "term"], ignore.case=TRUE)
cmpnd_Tf_df <- subset(cmpnd_Tf_df, !filter)
# Bug in tm_map(txt_corpus, removePunctuation, preserve_intra_word_dashes=TRUE) ???
# "net-a-porter" gets converted to "net-aporter"
#grep("net-a-porter", txt_vctr, ignore.case=TRUE, value=TRUE)
#grep("maser-laser", txt_vctr, ignore.case=TRUE, value=TRUE)
#txt_corpus[[which(grepl("net-a-porter", txt_vctr, ignore.case=TRUE))]]
#grep("\\b(across|longer)-(\\w)", cmpnd_Tf_df$term, ignore.case=TRUE, value=TRUE)
#grep("(\\w)-(affected|term)\\b", cmpnd_Tf_df$term, ignore.case=TRUE, value=TRUE)
print(sprintf("nrow(cmpnd_Tf_df): %d", nrow(cmpnd_Tf_df)))
myprint_df(cmpnd_Tf_df)
}
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "process.text_reporting_compound_terms"), major.inc=FALSE)
for (txt_var in glb_txt_vars) {
print(sprintf("Remaining compound terms in %s: ", txt_var))
txt_vctr <- glb_txt_lst[[txt_var]]
# find_cmpnd_wrds(txt_vctr)
#grep("thirty-five", txt_vctr, ignore.case=TRUE, value=TRUE)
#rex_str <- glb_txt_map_df[grepl("hirty", glb_txt_map_df$rex_str), "rex_str"]
}
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "build.corpus"), major.inc=TRUE)
glb_corpus_lst <- list()
print(sprintf("Building glb_corpus_lst..."))
glb_corpus_lst <- foreach(txt_var=glb_txt_vars) %dopar% {
# for (txt_var in glb_txt_vars) {
txt_corpus <- Corpus(VectorSource(glb_txt_lst[[txt_var]]))
txt_corpus <- tm_map(txt_corpus, tolower) #nuppr
txt_corpus <- tm_map(txt_corpus, PlainTextDocument)
txt_corpus <- tm_map(txt_corpus, removePunctuation) #npnct<chr_ix>
# txt-corpus <- tm_map(txt_corpus, content_transformer(function(x, pattern) gsub(pattern, "", x))
# Not to be run in production
inspect_terms <- function() {
full_Tf_DTM <- DocumentTermMatrix(txt_corpus,
control=list(weighting=weightTf))
print(" Full TermMatrix:"); print(full_Tf_DTM)
full_Tf_mtrx <- as.matrix(full_Tf_DTM)
rownames(full_Tf_mtrx) <- rownames(glb_allobs_df) # print undreadable otherwise
full_Tf_vctr <- colSums(full_Tf_mtrx)
names(full_Tf_vctr) <- dimnames(full_Tf_DTM)[[2]]
#grep("year", names(full_Tf_vctr), value=TRUE)
#which.max(full_Tf_mtrx[, "yearlong"])
full_Tf_df <- as.data.frame(full_Tf_vctr)
names(full_Tf_df) <- "Tf.full"
full_Tf_df$term <- rownames(full_Tf_df)
#full_Tf_df$freq.full <- colSums(full_Tf_mtrx != 0)
full_Tf_df <- orderBy(~ -Tf.full +term, full_Tf_df)
print(myplot_histogram(full_Tf_df, "Tf.full"))
myprint_df(full_Tf_df)
#txt_corpus[[which(grepl("zun", txt_vctr, ignore.case=TRUE))]]
digit_terms_df <- subset(full_Tf_df, grepl("[[:digit:]]", term))
myprint_df(digit_terms_df)
return(full_Tf_df)
}
#print("RemovePunct:"); remove_punct_Tf_df <- inspect_terms()
txt_corpus <- tm_map(txt_corpus, removeWords,
c(glb_append_stop_words[[txt_var]],
stopwords("english"))) #nstopwrds
#print("StoppedWords:"); stopped_words_Tf_df <- inspect_terms()
txt_corpus <- tm_map(txt_corpus, stemDocument) #Features for lost information: Difference/ratio in density of full_TfIdf_DTM ???
#txt_corpus <- tm_map(txt_corpus, content_transformer(stemDocument))
#print("StemmedWords:"); stemmed_words_Tf_df <- inspect_terms()
#stemmed_stopped_Tf_df <- merge(stemmed_words_Tf_df, stopped_words_Tf_df, by="term", all=TRUE, suffixes=c(".stem", ".stop"))
#myprint_df(stemmed_stopped_Tf_df)
#print(subset(stemmed_stopped_Tf_df, grepl("compan", term)))
#glb_corpus_lst[[txt_var]] <- txt_corpus
}
names(glb_corpus_lst) <- glb_txt_vars
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "extract.DTM"), major.inc=TRUE)
glb_full_DTM_lst <- list(); glb_sprs_DTM_lst <- list();
for (txt_var in glb_txt_vars) {
print(sprintf("Extracting TfIDf terms for %s...", txt_var))
txt_corpus <- glb_corpus_lst[[txt_var]]
# full_Tf_DTM <- DocumentTermMatrix(txt_corpus,
# control=list(weighting=weightTf))
full_TfIdf_DTM <- DocumentTermMatrix(txt_corpus,
control=list(weighting=weightTfIdf))
sprs_TfIdf_DTM <- removeSparseTerms(full_TfIdf_DTM,
glb_sprs_thresholds[txt_var])
# glb_full_DTM_lst[[txt_var]] <- full_Tf_DTM
# glb_sprs_DTM_lst[[txt_var]] <- sprs_Tf_DTM
glb_full_DTM_lst[[txt_var]] <- full_TfIdf_DTM
glb_sprs_DTM_lst[[txt_var]] <- sprs_TfIdf_DTM
}
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "report.DTM"), major.inc=TRUE)
for (txt_var in glb_txt_vars) {
print(sprintf("Reporting TfIDf terms for %s...", txt_var))
full_TfIdf_DTM <- glb_full_DTM_lst[[txt_var]]
sprs_TfIdf_DTM <- glb_sprs_DTM_lst[[txt_var]]
print(" Full TermMatrix:"); print(full_TfIdf_DTM)
full_TfIdf_mtrx <- as.matrix(full_TfIdf_DTM)
rownames(full_TfIdf_mtrx) <- rownames(glb_allobs_df) # print undreadable otherwise
full_TfIdf_vctr <- colSums(full_TfIdf_mtrx)
names(full_TfIdf_vctr) <- dimnames(full_TfIdf_DTM)[[2]]
#grep("scene", names(full_TfIdf_vctr), value=TRUE)
#which.max(full_TfIdf_mtrx[, "yearlong"])
full_TfIdf_df <- as.data.frame(full_TfIdf_vctr)
names(full_TfIdf_df) <- "TfIdf.full"
full_TfIdf_df$term <- rownames(full_TfIdf_df)
full_TfIdf_df$freq.full <- colSums(full_TfIdf_mtrx != 0)
full_TfIdf_df <- orderBy(~ -TfIdf.full, full_TfIdf_df)
print(" Sparse TermMatrix:"); print(sprs_TfIdf_DTM)
sprs_TfIdf_vctr <- colSums(as.matrix(sprs_TfIdf_DTM))
names(sprs_TfIdf_vctr) <- dimnames(sprs_TfIdf_DTM)[[2]]
sprs_TfIdf_df <- as.data.frame(sprs_TfIdf_vctr)
names(sprs_TfIdf_df) <- "TfIdf.sprs"
sprs_TfIdf_df$term <- rownames(sprs_TfIdf_df)
sprs_TfIdf_df$freq.sprs <- colSums(as.matrix(sprs_TfIdf_DTM) != 0)
sprs_TfIdf_df <- orderBy(~ -TfIdf.sprs, sprs_TfIdf_df)
terms_TfIdf_df <- merge(full_TfIdf_df, sprs_TfIdf_df, all.x=TRUE)
terms_TfIdf_df$in.sprs <- !is.na(terms_TfIdf_df$freq.sprs)
plt_TfIdf_df <- subset(terms_TfIdf_df,
TfIdf.full >= min(terms_TfIdf_df$TfIdf.sprs, na.rm=TRUE))
plt_TfIdf_df$label <- ""
plt_TfIdf_df[is.na(plt_TfIdf_df$TfIdf.sprs), "label"] <-
plt_TfIdf_df[is.na(plt_TfIdf_df$TfIdf.sprs), "term"]
glb_important_terms[[txt_var]] <- union(glb_important_terms[[txt_var]],
plt_TfIdf_df[is.na(plt_TfIdf_df$TfIdf.sprs), "term"])
print(myplot_scatter(plt_TfIdf_df, "freq.full", "TfIdf.full",
colorcol_name="in.sprs") +
geom_text(aes(label=label), color="Black", size=3.5))
melt_TfIdf_df <- orderBy(~ -value, melt(terms_TfIdf_df, id.var="term"))
print(ggplot(melt_TfIdf_df, aes(value, color=variable)) + stat_ecdf() +
geom_hline(yintercept=glb_sprs_thresholds[txt_var],
linetype = "dotted"))
melt_TfIdf_df <- orderBy(~ -value,
melt(subset(terms_TfIdf_df, !is.na(TfIdf.sprs)), id.var="term"))
print(myplot_hbar(melt_TfIdf_df, "term", "value",
colorcol_name="variable"))
melt_TfIdf_df <- orderBy(~ -value,
melt(subset(terms_TfIdf_df, is.na(TfIdf.sprs)), id.var="term"))
print(myplot_hbar(head(melt_TfIdf_df, 10), "term", "value",
colorcol_name="variable"))
}
# sav_full_DTM_lst <- glb_full_DTM_lst
# sav_sprs_DTM_lst <- glb_sprs_DTM_lst
# print(identical(sav_glb_corpus_lst, glb_corpus_lst))
# print(all.equal(length(sav_glb_corpus_lst), length(glb_corpus_lst)))
# print(all.equal(names(sav_glb_corpus_lst), names(glb_corpus_lst)))
# print(all.equal(sav_glb_corpus_lst[["Headline"]], glb_corpus_lst[["Headline"]]))
# print(identical(sav_full_DTM_lst, glb_full_DTM_lst))
# print(identical(sav_sprs_DTM_lst, glb_sprs_DTM_lst))
rm(full_TfIdf_mtrx, full_TfIdf_df, melt_TfIdf_df, terms_TfIdf_df)
# Create txt features
if ((length(glb_txt_vars) > 1) &&
(length(unique(pfxs <- sapply(glb_txt_vars,
function(txt) toupper(substr(txt, 1, 1))))) < length(glb_txt_vars)))
stop("Prefixes for corpus freq terms not unique: ", pfxs)
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "bind.DTM"),
major.inc=TRUE)
for (txt_var in glb_txt_vars) {
print(sprintf("Binding DTM for %s...", txt_var))
txt_var_pfx <- toupper(substr(txt_var, 1, 1))
txt_X_df <- as.data.frame(as.matrix(glb_sprs_DTM_lst[[txt_var]]))
colnames(txt_X_df) <- paste(txt_var_pfx, ".T.",
make.names(colnames(txt_X_df)), sep="")
rownames(txt_X_df) <- rownames(glb_allobs_df) # warning otherwise
# plt_X_df <- cbind(txt_X_df, glb_allobs_df[, c(glb_id_var, glb_rsp_var)])
# print(myplot_box(df=plt_X_df, ycol_names="H.T.today", xcol_name=glb_rsp_var))
# log_X_df <- log(1 + txt_X_df)
# colnames(log_X_df) <- paste(colnames(txt_X_df), ".log", sep="")
# plt_X_df <- cbind(log_X_df, glb_allobs_df[, c(glb_id_var, glb_rsp_var)])
# print(myplot_box(df=plt_X_df, ycol_names="H.T.today.log", xcol_name=glb_rsp_var))
glb_allobs_df <- cbind(glb_allobs_df, txt_X_df) # TfIdf is normalized
#glb_allobs_df <- cbind(glb_allobs_df, log_X_df) # if using non-normalized metrics
}
#identical(chk_entity_df, glb_allobs_df)
#chk_entity_df <- glb_allobs_df
extract.features_chunk_df <- myadd_chunk(extract.features_chunk_df,
paste0("extract.features_", "bind.DXM"),
major.inc=TRUE)
#sav_allobs_df <- glb_allobs_df
glb_punct_vctr <- c("!", "\"", "#", "\\$", "%", "&", "'",
"\\(|\\)",# "\\(", "\\)",
"\\*", "\\+", ",", "-", "\\.", "/", ":", ";",
"<|>", # "<",
"=",
# ">",
"\\?", "@", "\\[", "\\\\", "\\]", "^", "_", "`",
"\\{", "\\|", "\\}", "~")
txt_X_df <- glb_allobs_df[, c(glb_id_var, ".rnorm"), FALSE]
txt_X_df <- foreach(txt_var=glb_txt_vars, .combine=cbind) %dopar% {
#for (txt_var in glb_txt_vars) {
print(sprintf("Binding DXM for %s...", txt_var))
txt_var_pfx <- toupper(substr(txt_var, 1, 1))
#txt_X_df <- glb_allobs_df[, c(glb_id_var, ".rnorm"), FALSE]
txt_full_DTM_mtrx <- as.matrix(glb_full_DTM_lst[[txt_var]])
rownames(txt_full_DTM_mtrx) <- rownames(glb_allobs_df) # print undreadable otherwise
#print(txt_full_DTM_mtrx[txt_full_DTM_mtrx[, "ebola"] != 0, "ebola"])
# Create <txt_var>.T.<term> for glb_important_terms
for (term in glb_important_terms[[txt_var]])
txt_X_df[, paste0(txt_var_pfx, ".T.", make.names(term))] <-
txt_full_DTM_mtrx[, term]
# Create <txt_var>.nwrds.log & .nwrds.unq.log
txt_X_df[, paste0(txt_var_pfx, ".nwrds.log")] <-
log(1 + mycount_pattern_occ("\\w+", glb_txt_lst[[txt_var]]))
txt_X_df[, paste0(txt_var_pfx, ".nwrds.unq.log")] <-
log(1 + rowSums(txt_full_DTM_mtrx != 0))
txt_X_df[, paste0(txt_var_pfx, ".sum.TfIdf")] <-
rowSums(txt_full_DTM_mtrx)
txt_X_df[, paste0(txt_var_pfx, ".ratio.sum.TfIdf.nwrds")] <-
txt_X_df[, paste0(txt_var_pfx, ".sum.TfIdf")] /
(exp(txt_X_df[, paste0(txt_var_pfx, ".nwrds.log")]) - 1)
txt_X_df[is.nan(txt_X_df[, paste0(txt_var_pfx, ".ratio.sum.TfIdf.nwrds")]),
paste0(txt_var_pfx, ".ratio.sum.TfIdf.nwrds")] <- 0
# Create <txt_var>.nchrs.log
txt_X_df[, paste0(txt_var_pfx, ".nchrs.log")] <-
log(1 + mycount_pattern_occ(".", glb_allobs_df[, txt_var]))
txt_X_df[, paste0(txt_var_pfx, ".nuppr.log")] <-
log(1 + mycount_pattern_occ("[[:upper:]]", glb_allobs_df[, txt_var]))
txt_X_df[, paste0(txt_var_pfx, ".ndgts.log")] <-
log(1 + mycount_pattern_occ("[[:digit:]]", glb_allobs_df[, txt_var]))
# Create <txt_var>.npnct?.log
# would this be faster if it's iterated over each row instead of
# each created column ???
for (punct_ix in 1:length(glb_punct_vctr)) {
# smp0 <- " "
# smp1 <- "! \" # $ % & ' ( ) * + , - . / : ; < = > ? @ [ \ ] ^ _ ` { | } ~"
# smp2 <- paste(smp1, smp1, sep=" ")
# print(sprintf("Testing %s pattern:", glb_punct_vctr[punct_ix]))
# results <- mycount_pattern_occ(glb_punct_vctr[punct_ix], c(smp0, smp1, smp2))
# names(results) <- NULL; print(results)
txt_X_df[,
paste0(txt_var_pfx, ".npnct", sprintf("%02d", punct_ix), ".log")] <-
log(1 + mycount_pattern_occ(glb_punct_vctr[punct_ix],
glb_allobs_df[, txt_var]))
}
# print(head(glb_allobs_df[glb_allobs_df[, "A.npnct23.log"] > 0,
# c("UniqueID", "Popular", "Abstract", "A.npnct23.log")]))
# Create <txt_var>.nstopwrds.log & <txt_var>ratio.nstopwrds.nwrds
stop_words_rex_str <- paste0("\\b(", paste0(c(glb_append_stop_words[[txt_var]],
stopwords("english")), collapse="|"),
")\\b")
txt_X_df[, paste0(txt_var_pfx, ".nstopwrds", ".log")] <-
log(1 + mycount_pattern_occ(stop_words_rex_str, glb_txt_lst[[txt_var]]))
txt_X_df[, paste0(txt_var_pfx, ".ratio.nstopwrds.nwrds")] <-
exp(txt_X_df[, paste0(txt_var_pfx, ".nstopwrds", ".log")] -
txt_X_df[, paste0(txt_var_pfx, ".nwrds", ".log")])
# Create <txt_var>.P.http
txt_X_df[, paste(txt_var_pfx, ".P.http", sep="")] <-
as.integer(0 + mycount_pattern_occ("http", glb_allobs_df[, txt_var]))
txt_X_df <- subset(txt_X_df, select=-.rnorm)
txt_X_df <- txt_X_df[, -grep(glb_id_var, names(txt_X_df), fixed=TRUE), FALSE]
#glb_allobs_df <- cbind(glb_allobs_df, txt_X_df)
}
glb_allobs_df <- cbind(glb_allobs_df, txt_X_df)
#myplot_box(glb_allobs_df, "A.sum.TfIdf", glb_rsp_var)
# Generate summaries
# print(summary(glb_allobs_df))
# print(sapply(names(glb_allobs_df), function(col) sum(is.na(glb_allobs_df[, col]))))
# print(summary(glb_trnobs_df))
# print(sapply(names(glb_trnobs_df), function(col) sum(is.na(glb_trnobs_df[, col]))))
# print(summary(glb_newobs_df))
# print(sapply(names(glb_newobs_df), function(col) sum(is.na(glb_newobs_df[, col]))))
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_txt_vars)
rm(log_X_df, txt_X_df)
}
# print(sapply(names(glb_trnobs_df), function(col) sum(is.na(glb_trnobs_df[, col]))))
# print(sapply(names(glb_newobs_df), function(col) sum(is.na(glb_newobs_df[, col]))))
# print(myplot_scatter(glb_trnobs_df, "<col1_name>", "<col2_name>", smooth=TRUE))
rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
glb_full_DTM_lst, glb_sprs_DTM_lst, txt_corpus, txt_vctr)## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'corpus_lst' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'full_TfIdf_DTM' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'full_TfIdf_vctr' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'glb_full_DTM_lst' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'glb_sprs_DTM_lst' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'txt_corpus' not found## Warning in rm(corpus_lst, full_TfIdf_DTM, full_TfIdf_vctr,
## glb_full_DTM_lst, : object 'txt_vctr' not foundextract.features_chunk_df <- myadd_chunk(extract.features_chunk_df, "extract.features_end",
major.inc=TRUE)## label step_major step_minor bgn end
## 2 extract.features_factorize.str.vars 2 0 31.409 31.428
## 3 extract.features_end 3 0 31.428 NA
## elapsed
## 2 0.019
## 3 NAmyplt_chunk(extract.features_chunk_df)## label step_major step_minor bgn end
## 2 extract.features_factorize.str.vars 2 0 31.409 31.428
## 1 extract.features_bgn 1 0 31.395 31.409
## elapsed duration
## 2 0.019 0.019
## 1 0.014 0.014
## [1] "Total Elapsed Time: 31.428 secs"
# if (glb_save_envir)
# save(glb_feats_df,
# glb_allobs_df, #glb_trnobs_df, glb_fitobs_df, glb_OOBobs_df, glb_newobs_df,
# file=paste0(glb_out_pfx, "extract_features_dsk.RData"))
# load(paste0(glb_out_pfx, "extract_features_dsk.RData"))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all","data.new")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
glb_chunks_df <- myadd_chunk(glb_chunks_df, "cluster.data", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 5 extract.features 3 0 31.389 32.797 1.408
## 6 cluster.data 4 0 32.798 NA NAStep 4.0: cluster data
glb_chunks_df <- myadd_chunk(glb_chunks_df, "manage.missing.data", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 6 cluster.data 4 0 32.798 33.095 0.298
## 7 manage.missing.data 4 1 33.096 NA NA# If mice crashes with error: Error in get(as.character(FUN), mode = "function", envir = envir) : object 'State' of mode 'function' was not found
# consider excluding 'State' as a feature
# print(sapply(names(glb_trnobs_df), function(col) sum(is.na(glb_trnobs_df[, col]))))
# print(sapply(names(glb_newobs_df), function(col) sum(is.na(glb_newobs_df[, col]))))
# glb_trnobs_df <- na.omit(glb_trnobs_df)
# glb_newobs_df <- na.omit(glb_newobs_df)
# df[is.na(df)] <- 0
mycheck_problem_data(glb_allobs_df)## [1] "numeric data missing in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ 0s in glb_allobs_df: "
## timesignature timesignature_confidence tempo
## 10 157 13
## tempo_confidence key key_confidence
## 85 803 422
## pitch timbre_0_min Top10
## 108 3114 6455
## [1] "numeric data w/ Infs in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ NaNs in glb_allobs_df: "
## named integer(0)
## [1] "string data missing in glb_allobs_df: "
## songtitle artistname songID artistID .rownames
## 0 0 0 0 0# glb_allobs_df <- na.omit(glb_allobs_df)
# Not refactored into mydsutils.R since glb_*_df might be reassigned
glb_impute_missing_data <- function() {
require(mice)
set.seed(glb_mice_complete.seed)
inp_impent_df <- glb_allobs_df[, setdiff(names(glb_allobs_df),
union(glb_exclude_vars_as_features, glb_rsp_var))]
print("Summary before imputation: ")
print(summary(inp_impent_df))
out_impent_df <- complete(mice(inp_impent_df))
print(summary(out_impent_df))
ret_vars <- sapply(names(out_impent_df),
function(col) ifelse(!identical(out_impent_df[, col],
inp_impent_df[, col]),
col, ""))
ret_vars <- ret_vars[ret_vars != ""]
# complete(mice()) changes attributes of factors even though values don't change
for (col in ret_vars) {
if (inherits(out_impent_df[, col], "factor")) {
if (identical(as.numeric(out_impent_df[, col]),
as.numeric(inp_impent_df[, col])))
ret_vars <- setdiff(ret_vars, col)
}
}
return(out_impent_df[, ret_vars])
}
if (glb_impute_na_data &&
(length(myfind_numerics_missing(glb_allobs_df)) > 0) &&
(ncol(nonna_df <- glb_impute_missing_data()) > 0)) {
for (col in names(nonna_df)) {
glb_allobs_df[, paste0(col, ".nonNA")] <- nonna_df[, col]
glb_exclude_vars_as_features <- c(glb_exclude_vars_as_features, col)
}
}
mycheck_problem_data(glb_allobs_df, terminate = TRUE)## [1] "numeric data missing in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ 0s in glb_allobs_df: "
## timesignature timesignature_confidence tempo
## 10 157 13
## tempo_confidence key key_confidence
## 85 803 422
## pitch timbre_0_min Top10
## 108 3114 6455
## [1] "numeric data w/ Infs in glb_allobs_df: "
## named integer(0)
## [1] "numeric data w/ NaNs in glb_allobs_df: "
## named integer(0)
## [1] "string data missing in glb_allobs_df: "
## songtitle artistname songID artistID .rownames
## 0 0 0 0 0Step 4.1: manage missing data
if (glb_cluster) {
require(proxy)
#require(hash)
require(dynamicTreeCut)
# glb_hash <- hash(key=unique(glb_allobs_df$myCategory),
# values=1:length(unique(glb_allobs_df$myCategory)))
# glb_hash_lst <- hash(key=unique(glb_allobs_df$myCategory),
# values=1:length(unique(glb_allobs_df$myCategory)))
#stophere; sav_allobs_df <- glb_allobs_df;
print("Clustering features: ")
print(cluster_vars <- grep("[HSA]\\.[PT]\\.", names(glb_allobs_df), value=TRUE))
#print(cluster_vars <- grep("[HSA]\\.", names(glb_allobs_df), value=TRUE))
glb_allobs_df$.clusterid <- 1
#print(max(table(glb_allobs_df$myCategory.fctr) / 20))
for (myCategory in c("##", "Business#Business Day#Dealbook", "OpEd#Opinion#",
"Styles#U.S.#", "Business#Technology#", "Science#Health#",
"Culture#Arts#")) {
ctgry_allobs_df <- glb_allobs_df[glb_allobs_df$myCategory == myCategory, ]
dstns_dist <- dist(ctgry_allobs_df[, cluster_vars], method = "cosine")
dstns_mtrx <- as.matrix(dstns_dist)
print(sprintf("max distance(%0.4f) pair:", max(dstns_mtrx)))
row_ix <- ceiling(which.max(dstns_mtrx) / ncol(dstns_mtrx))
col_ix <- which.max(dstns_mtrx[row_ix, ])
print(ctgry_allobs_df[c(row_ix, col_ix),
c("UniqueID", "Popular", "myCategory", "Headline", cluster_vars)])
min_dstns_mtrx <- dstns_mtrx
diag(min_dstns_mtrx) <- 1
print(sprintf("min distance(%0.4f) pair:", min(min_dstns_mtrx)))
row_ix <- ceiling(which.min(min_dstns_mtrx) / ncol(min_dstns_mtrx))
col_ix <- which.min(min_dstns_mtrx[row_ix, ])
print(ctgry_allobs_df[c(row_ix, col_ix),
c("UniqueID", "Popular", "myCategory", "Headline", cluster_vars)])
clusters <- hclust(dstns_dist, method = "ward.D2")
#plot(clusters, labels=NULL, hang=-1)
myplclust(clusters, lab.col=unclass(ctgry_allobs_df[, glb_rsp_var]))
#clusterGroups = cutree(clusters, k=7)
clusterGroups <- cutreeDynamic(clusters, minClusterSize=20, method="tree", deepSplit=0)
# Unassigned groups are labeled 0; the largest group has label 1
table(clusterGroups, ctgry_allobs_df[, glb_rsp_var], useNA="ifany")
#print(ctgry_allobs_df[which(clusterGroups == 1), c("UniqueID", "Popular", "Headline")])
#print(ctgry_allobs_df[(clusterGroups == 1) & !is.na(ctgry_allobs_df$Popular) & (ctgry_allobs_df$Popular==1), c("UniqueID", "Popular", "Headline")])
clusterGroups[clusterGroups == 0] <- 1
table(clusterGroups, ctgry_allobs_df[, glb_rsp_var], useNA="ifany")
#summary(factor(clusterGroups))
# clusterGroups <- clusterGroups +
# 100 * # has to be > max(table(glb_allobs_df$myCategory.fctr) / minClusterSize=20)
# which(levels(glb_allobs_df$myCategory.fctr) == myCategory)
# table(clusterGroups, ctgry_allobs_df[, glb_rsp_var], useNA="ifany")
# add to glb_allobs_df - then split the data again
glb_allobs_df[glb_allobs_df$myCategory==myCategory,]$.clusterid <- clusterGroups
#print(unique(glb_allobs_df$.clusterid))
#print(glb_feats_df[glb_feats_df$id == ".clusterid.fctr", ])
}
ctgry_xtab_df <- orderBy(reformulate(c("-", ".n")),
mycreate_sqlxtab_df(glb_allobs_df,
c("myCategory", ".clusterid", glb_rsp_var)))
ctgry_cast_df <- orderBy(~ -Y -NA, dcast(ctgry_xtab_df,
myCategory + .clusterid ~
Popular.fctr, sum, value.var=".n"))
print(ctgry_cast_df)
#print(orderBy(~ myCategory -Y -NA, ctgry_cast_df))
# write.table(ctgry_cast_df, paste0(glb_out_pfx, "ctgry_clst.csv"),
# row.names=FALSE)
print(ctgry_sum_tbl <- table(glb_allobs_df$myCategory, glb_allobs_df$.clusterid,
glb_allobs_df[, glb_rsp_var],
useNA="ifany"))
# dsp_obs(.clusterid=1, myCategory="OpEd#Opinion#",
# cols=c("UniqueID", "Popular", "myCategory", ".clusterid", "Headline"),
# all=TRUE)
glb_allobs_df$.clusterid.fctr <- as.factor(glb_allobs_df$.clusterid)
glb_exclude_vars_as_features <- c(glb_exclude_vars_as_features,
".clusterid")
glb_interaction_only_features["myCategory.fctr"] <- c(".clusterid.fctr")
glb_exclude_vars_as_features <- c(glb_exclude_vars_as_features,
cluster_vars)
}
# Last call for data modifications
#stop(here") # sav_allobs_df <- glb_allobs_df
# glb_allobs_df[(glb_allobs_df$PropR == 0.75) & (glb_allobs_df$State == "Hawaii"), "PropR.fctr"] <- "N"
# Re-partition
glb_trnobs_df <- subset(glb_allobs_df, .src == "Train")
glb_newobs_df <- subset(glb_allobs_df, .src == "Test")
glb_chunks_df <- myadd_chunk(glb_chunks_df, "select.features", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 7 manage.missing.data 4 1 33.096 33.219 0.124
## 8 select.features 5 0 33.220 NA NAStep 5.0: select features
print(glb_feats_df <- myselect_features(entity_df=glb_trnobs_df,
exclude_vars_as_features=glb_exclude_vars_as_features,
rsp_var=glb_rsp_var))## id cor.y
## Top10 Top10 1.000000000
## timbre_6_min timbre_6_min -0.216378268
## timbre_4_max timbre_4_max 0.197034666
## timbre_11_min timbre_11_min -0.193049686
## timbre_8_max timbre_8_max 0.162722687
## timbre_7_min timbre_7_min -0.155645339
## timbre_0_max timbre_0_max -0.152171355
## timbre_9_max timbre_9_max 0.145756048
## timbre_11_max timbre_11_max 0.143718363
## timbre_2_min timbre_2_min -0.136085935
## pitch pitch -0.135723510
## timbre_5_min timbre_5_min -0.123590779
## timbre_6_max timbre_6_max 0.123432076
## energy energy -0.120897670
## year year -0.106988968
## timbre_1_max timbre_1_max 0.106338252
## timbre_10_max timbre_10_max 0.104441775
## timbre_8_min timbre_8_min -0.101131084
## timbre_2_max timbre_2_max 0.096545551
## timbre_9_min timbre_9_min -0.088167503
## loudness loudness -0.086546031
## tempo_confidence tempo_confidence 0.086315020
## timbre_5_max timbre_5_max 0.076923330
## timbre_0_min timbre_0_min 0.068825164
## timesignature_confidence timesignature_confidence 0.063023674
## timesignature timesignature 0.046630910
## timbre_7_max timbre_7_max 0.045966630
## timbre_3_max timbre_3_max -0.026744876
## timbre_3_min timbre_3_min -0.026442066
## key key 0.025326063
## .rnorm .rnorm 0.011773171
## key_confidence key_confidence 0.010021403
## timbre_4_min timbre_4_min -0.006010064
## timbre_10_min timbre_10_min 0.005257590
## timbre_1_min timbre_1_min 0.003659089
## tempo tempo -0.003577709
## exclude.as.feat cor.y.abs
## Top10 1 1.000000000
## timbre_6_min 0 0.216378268
## timbre_4_max 0 0.197034666
## timbre_11_min 0 0.193049686
## timbre_8_max 0 0.162722687
## timbre_7_min 0 0.155645339
## timbre_0_max 0 0.152171355
## timbre_9_max 0 0.145756048
## timbre_11_max 0 0.143718363
## timbre_2_min 0 0.136085935
## pitch 0 0.135723510
## timbre_5_min 0 0.123590779
## timbre_6_max 0 0.123432076
## energy 0 0.120897670
## year 1 0.106988968
## timbre_1_max 0 0.106338252
## timbre_10_max 0 0.104441775
## timbre_8_min 0 0.101131084
## timbre_2_max 0 0.096545551
## timbre_9_min 0 0.088167503
## loudness 0 0.086546031
## tempo_confidence 0 0.086315020
## timbre_5_max 0 0.076923330
## timbre_0_min 0 0.068825164
## timesignature_confidence 0 0.063023674
## timesignature 0 0.046630910
## timbre_7_max 0 0.045966630
## timbre_3_max 0 0.026744876
## timbre_3_min 0 0.026442066
## key 0 0.025326063
## .rnorm 0 0.011773171
## key_confidence 0 0.010021403
## timbre_4_min 0 0.006010064
## timbre_10_min 0 0.005257590
## timbre_1_min 0 0.003659089
## tempo 0 0.003577709# sav_feats_df <- glb_feats_df; glb_feats_df <- sav_feats_df
print(glb_feats_df <- orderBy(~-cor.y,
myfind_cor_features(feats_df=glb_feats_df, obs_df=glb_trnobs_df,
rsp_var=glb_rsp_var)))## [1] "cor(loudness, timbre_0_max)=0.9059"
## [1] "cor(Top10.fctr, loudness)=-0.0865"
## [1] "cor(Top10.fctr, timbre_0_max)=-0.1522"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glb_trnobs_df, : Identified loudness as highly correlated with timbre_0_max## id cor.y exclude.as.feat cor.y.abs
## 35 Top10 1.000000000 1 1.000000000
## 21 timbre_4_max 0.197034666 0 0.197034666
## 29 timbre_8_max 0.162722687 0 0.162722687
## 31 timbre_9_max 0.145756048 0 0.145756048
## 15 timbre_11_max 0.143718363 0 0.143718363
## 25 timbre_6_max 0.123432076 0 0.123432076
## 11 timbre_1_max 0.106338252 0 0.106338252
## 13 timbre_10_max 0.104441775 0 0.104441775
## 17 timbre_2_max 0.096545551 0 0.096545551
## 8 tempo_confidence 0.086315020 0 0.086315020
## 23 timbre_5_max 0.076923330 0 0.076923330
## 10 timbre_0_min 0.068825164 0 0.068825164
## 34 timesignature_confidence 0.063023674 0 0.063023674
## 33 timesignature 0.046630910 0 0.046630910
## 27 timbre_7_max 0.045966630 0 0.045966630
## 3 key 0.025326063 0 0.025326063
## 1 .rnorm 0.011773171 0 0.011773171
## 4 key_confidence 0.010021403 0 0.010021403
## 14 timbre_10_min 0.005257590 0 0.005257590
## 12 timbre_1_min 0.003659089 0 0.003659089
## 7 tempo -0.003577709 0 0.003577709
## 22 timbre_4_min -0.006010064 0 0.006010064
## 20 timbre_3_min -0.026442066 0 0.026442066
## 19 timbre_3_max -0.026744876 0 0.026744876
## 5 loudness -0.086546031 0 0.086546031
## 32 timbre_9_min -0.088167503 0 0.088167503
## 30 timbre_8_min -0.101131084 0 0.101131084
## 36 year -0.106988968 1 0.106988968
## 2 energy -0.120897670 0 0.120897670
## 24 timbre_5_min -0.123590779 0 0.123590779
## 6 pitch -0.135723510 0 0.135723510
## 18 timbre_2_min -0.136085935 0 0.136085935
## 9 timbre_0_max -0.152171355 0 0.152171355
## 28 timbre_7_min -0.155645339 0 0.155645339
## 16 timbre_11_min -0.193049686 0 0.193049686
## 26 timbre_6_min -0.216378268 0 0.216378268
## cor.high.X freqRatio percentUnique zeroVar nzv myNearZV
## 35 <NA> 5.793396 0.02777392 FALSE FALSE FALSE
## 21 <NA> 80.333333 93.51478961 FALSE FALSE FALSE
## 29 <NA> 1.250000 92.32051104 FALSE FALSE FALSE
## 31 <NA> 1.000000 95.04235523 FALSE FALSE FALSE
## 15 <NA> 1.333333 91.86224136 FALSE FALSE FALSE
## 25 <NA> 1.000000 94.11192890 FALSE FALSE FALSE
## 11 <NA> 126.571429 84.90487432 FALSE FALSE FALSE
## 13 <NA> 1.000000 94.55631162 FALSE FALSE FALSE
## 17 <NA> 1.000000 97.51423413 FALSE FALSE FALSE
## 8 <NA> 6.554217 13.67865574 FALSE FALSE FALSE
## 23 <NA> 1.000000 97.97250382 FALSE FALSE FALSE
## 10 <NA> 21.099291 37.77253159 FALSE FALSE FALSE
## 34 <NA> 18.928571 11.22066380 FALSE FALSE FALSE
## 33 <NA> 13.717622 0.08332176 FALSE FALSE FALSE
## 27 <NA> 1.000000 96.05610332 FALSE FALSE FALSE
## 3 <NA> 1.052696 0.16664352 FALSE FALSE FALSE
## 1 <NA> 1.000000 100.00000000 FALSE FALSE FALSE
## 4 <NA> 4.093750 13.59533398 FALSE FALSE FALSE
## 14 <NA> 1.000000 95.82002500 FALSE FALSE FALSE
## 12 <NA> 1.000000 98.23635606 FALSE FALSE FALSE
## 7 <NA> 2.600000 92.90376337 FALSE FALSE FALSE
## 22 <NA> 1.333333 94.72295515 FALSE FALSE FALSE
## 20 <NA> 1.000000 98.09748646 FALSE FALSE FALSE
## 19 <NA> 1.000000 98.55575614 FALSE FALSE FALSE
## 5 timbre_0_max 1.000000 76.30884599 FALSE FALSE FALSE
## 32 <NA> 1.000000 94.08415498 FALSE FALSE FALSE
## 30 <NA> 5.500000 93.05651993 FALSE FALSE FALSE
## 36 <NA> 1.200772 0.27773920 FALSE FALSE FALSE
## 2 <NA> 1.000000 99.63893904 FALSE FALSE FALSE
## 24 <NA> 2.250000 95.25065963 FALSE FALSE FALSE
## 6 <NA> 1.007987 1.33314817 FALSE FALSE FALSE
## 18 <NA> 1.666667 97.62532982 FALSE FALSE FALSE
## 9 <NA> 1.166667 72.68434940 FALSE FALSE FALSE
## 28 <NA> 1.000000 95.88945980 FALSE FALSE FALSE
## 16 <NA> 1.000000 92.04277184 FALSE FALSE FALSE
## 26 <NA> 1.000000 95.16733787 FALSE FALSE FALSE
## is.cor.y.abs.low
## 35 FALSE
## 21 FALSE
## 29 FALSE
## 31 FALSE
## 15 FALSE
## 25 FALSE
## 11 FALSE
## 13 FALSE
## 17 FALSE
## 8 FALSE
## 23 FALSE
## 10 FALSE
## 34 FALSE
## 33 FALSE
## 27 FALSE
## 3 FALSE
## 1 FALSE
## 4 TRUE
## 14 TRUE
## 12 TRUE
## 7 TRUE
## 22 TRUE
## 20 FALSE
## 19 FALSE
## 5 FALSE
## 32 FALSE
## 30 FALSE
## 36 FALSE
## 2 FALSE
## 24 FALSE
## 6 FALSE
## 18 FALSE
## 9 FALSE
## 28 FALSE
## 16 FALSE
## 26 FALSE#subset(glb_feats_df, id %in% c("A.nuppr.log", "S.nuppr.log"))
print(myplot_scatter(glb_feats_df, "percentUnique", "freqRatio",
colorcol_name="myNearZV", jitter=TRUE) +
geom_point(aes(shape=nzv)) + xlim(-5, 25))## Warning in myplot_scatter(glb_feats_df, "percentUnique", "freqRatio",
## colorcol_name = "myNearZV", : converting myNearZV to class:factor## Warning: Removed 28 rows containing missing values (geom_point).## Warning: Removed 28 rows containing missing values (geom_point).## Warning: Removed 28 rows containing missing values (geom_point).
print(subset(glb_feats_df, myNearZV))## [1] id cor.y exclude.as.feat cor.y.abs
## [5] cor.high.X freqRatio percentUnique zeroVar
## [9] nzv myNearZV is.cor.y.abs.low
## <0 rows> (or 0-length row.names)glb_allobs_df <- glb_allobs_df[, setdiff(names(glb_allobs_df),
subset(glb_feats_df, myNearZV)$id)]
if (!is.null(glb_interaction_only_features))
glb_feats_df[glb_feats_df$id %in% glb_interaction_only_features, "interaction.feat"] <-
names(glb_interaction_only_features) else
glb_feats_df$interaction.feat <- NA
mycheck_problem_data(glb_allobs_df, terminate = TRUE)## [1] "numeric data missing in : "
## named integer(0)
## [1] "numeric data w/ 0s in : "
## timesignature timesignature_confidence tempo
## 10 157 13
## tempo_confidence key key_confidence
## 85 803 422
## pitch timbre_0_min Top10
## 108 3114 6455
## [1] "numeric data w/ Infs in : "
## named integer(0)
## [1] "numeric data w/ NaNs in : "
## named integer(0)
## [1] "string data missing in : "
## songtitle artistname songID artistID .rownames
## 0 0 0 0 0# glb_allobs_df %>% filter(is.na(Married.fctr)) %>% tbl_df()
# glb_allobs_df %>% count(Married.fctr)
# levels(glb_allobs_df$Married.fctr)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "partition.data.training", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 8 select.features 5 0 33.220 35.356 2.136
## 9 partition.data.training 6 0 35.356 NA NAStep 6.0: partition data training
if (all(is.na(glb_newobs_df[, glb_rsp_var]))) {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnobs_df[, glb_rsp_var_raw],
SplitRatio=1 - (nrow(glb_newobs_df) * 1.1 / nrow(glb_trnobs_df)))
glb_fitobs_df <- glb_trnobs_df[split, ]
glb_OOBobs_df <- glb_trnobs_df[!split ,]
} else {
print(sprintf("Newdata contains non-NA data for %s; setting OOB to Newdata",
glb_rsp_var))
glb_fitobs_df <- glb_trnobs_df; glb_OOBobs_df <- glb_newobs_df
}## [1] "Newdata contains non-NA data for Top10.fctr; setting OOB to Newdata"if (!is.null(glb_max_fitobs) && (nrow(glb_fitobs_df) > glb_max_fitobs)) {
warning("glb_fitobs_df restricted to glb_max_fitobs: ",
format(glb_max_fitobs, big.mark=","))
org_fitobs_df <- glb_fitobs_df
glb_fitobs_df <-
org_fitobs_df[split <- sample.split(org_fitobs_df[, glb_rsp_var_raw],
SplitRatio=glb_max_fitobs), ]
org_fitobs_df <- NULL
}
glb_allobs_df$.lcn <- ""
glb_allobs_df[glb_allobs_df[, glb_id_var] %in%
glb_fitobs_df[, glb_id_var], ".lcn"] <- "Fit"
glb_allobs_df[glb_allobs_df[, glb_id_var] %in%
glb_OOBobs_df[, glb_id_var], ".lcn"] <- "OOB"
dsp_class_dstrb <- function(obs_df, location_var, partition_var) {
xtab_df <- mycreate_xtab_df(obs_df, c(location_var, partition_var))
rownames(xtab_df) <- xtab_df[, location_var]
xtab_df <- xtab_df[, -grepl(location_var, names(xtab_df))]
print(xtab_df)
print(xtab_df / rowSums(xtab_df, na.rm=TRUE))
}
# Ensure proper splits by glb_rsp_var_raw & user-specified feature for OOB vs. new
if (!is.null(glb_category_vars)) {
if (glb_is_classification)
dsp_class_dstrb(glb_allobs_df, ".lcn", glb_rsp_var_raw)
newobs_ctgry_df <- mycreate_sqlxtab_df(subset(glb_allobs_df, .src == "Test"),
glb_category_vars)
OOBobs_ctgry_df <- mycreate_sqlxtab_df(subset(glb_allobs_df, .lcn == "OOB"),
glb_category_vars)
glb_ctgry_df <- merge(newobs_ctgry_df, OOBobs_ctgry_df, by=glb_category_vars
, all=TRUE, suffixes=c(".Tst", ".OOB"))
glb_ctgry_df$.freqRatio.Tst <- glb_ctgry_df$.n.Tst / sum(glb_ctgry_df$.n.Tst, na.rm=TRUE)
glb_ctgry_df$.freqRatio.OOB <- glb_ctgry_df$.n.OOB / sum(glb_ctgry_df$.n.OOB, na.rm=TRUE)
print(orderBy(~-.freqRatio.Tst-.freqRatio.OOB, glb_ctgry_df))
}
# Run this line by line
print("glb_feats_df:"); print(dim(glb_feats_df))## [1] "glb_feats_df:"## [1] 36 12sav_feats_df <- glb_feats_df
glb_feats_df <- sav_feats_df
glb_feats_df[, "rsp_var_raw"] <- FALSE
glb_feats_df[glb_feats_df$id == glb_rsp_var_raw, "rsp_var_raw"] <- TRUE
glb_feats_df$exclude.as.feat <- (glb_feats_df$exclude.as.feat == 1)
if (!is.null(glb_id_var) && glb_id_var != ".rownames")
glb_feats_df[glb_feats_df$id %in% glb_id_var, "id_var"] <- TRUE
add_feats_df <- data.frame(id=glb_rsp_var, exclude.as.feat=TRUE, rsp_var=TRUE)
row.names(add_feats_df) <- add_feats_df$id; print(add_feats_df)## id exclude.as.feat rsp_var
## Top10.fctr Top10.fctr TRUE TRUEglb_feats_df <- myrbind_df(glb_feats_df, add_feats_df)
if (glb_id_var != ".rownames")
print(subset(glb_feats_df, rsp_var_raw | rsp_var | id_var)) else
print(subset(glb_feats_df, rsp_var_raw | rsp_var)) ## id cor.y exclude.as.feat cor.y.abs cor.high.X freqRatio
## 35 Top10 1 TRUE 1 <NA> 5.793396
## Top10.fctr Top10.fctr NA TRUE NA <NA> NA
## percentUnique zeroVar nzv myNearZV is.cor.y.abs.low
## 35 0.02777392 FALSE FALSE FALSE FALSE
## Top10.fctr NA NA NA NA NA
## interaction.feat rsp_var_raw rsp_var
## 35 NA TRUE NA
## Top10.fctr NA NA TRUEprint("glb_feats_df vs. glb_allobs_df: "); ## [1] "glb_feats_df vs. glb_allobs_df: "print(setdiff(glb_feats_df$id, names(glb_allobs_df)))## character(0)print("glb_allobs_df vs. glb_feats_df: "); ## [1] "glb_allobs_df vs. glb_feats_df: "# Ensure these are only chr vars
print(setdiff(setdiff(names(glb_allobs_df), glb_feats_df$id),
myfind_chr_cols_df(glb_allobs_df)))## character(0)#print(setdiff(setdiff(names(glb_allobs_df), glb_exclude_vars_as_features),
# glb_feats_df$id))
print("glb_allobs_df: "); print(dim(glb_allobs_df))## [1] "glb_allobs_df: "## [1] 7574 44print("glb_trnobs_df: "); print(dim(glb_trnobs_df))## [1] "glb_trnobs_df: "## [1] 7201 43print("glb_fitobs_df: "); print(dim(glb_fitobs_df))## [1] "glb_fitobs_df: "## [1] 7201 43print("glb_OOBobs_df: "); print(dim(glb_OOBobs_df))## [1] "glb_OOBobs_df: "## [1] 373 43print("glb_newobs_df: "); print(dim(glb_newobs_df))## [1] "glb_newobs_df: "## [1] 373 43# # Does not handle NULL or length(glb_id_var) > 1
# glb_allobs_df$.src.trn <- 0
# glb_allobs_df[glb_allobs_df[, glb_id_var] %in% glb_trnobs_df[, glb_id_var],
# ".src.trn"] <- 1
# glb_allobs_df$.src.fit <- 0
# glb_allobs_df[glb_allobs_df[, glb_id_var] %in% glb_fitobs_df[, glb_id_var],
# ".src.fit"] <- 1
# glb_allobs_df$.src.OOB <- 0
# glb_allobs_df[glb_allobs_df[, glb_id_var] %in% glb_OOBobs_df[, glb_id_var],
# ".src.OOB"] <- 1
# glb_allobs_df$.src.new <- 0
# glb_allobs_df[glb_allobs_df[, glb_id_var] %in% glb_newobs_df[, glb_id_var],
# ".src.new"] <- 1
# #print(unique(glb_allobs_df[, ".src.trn"]))
# write_cols <- c(glb_feats_df$id,
# ".src.trn", ".src.fit", ".src.OOB", ".src.new")
# glb_allobs_df <- glb_allobs_df[, write_cols]
#
# tmp_feats_df <- glb_feats_df
# tmp_entity_df <- glb_allobs_df
if (glb_save_envir)
save(glb_feats_df,
glb_allobs_df, #glb_trnobs_df, glb_fitobs_df, glb_OOBobs_df, glb_newobs_df,
file=paste0(glb_out_pfx, "blddfs_dsk.RData"))
# load(paste0(glb_out_pfx, "blddfs_dsk.RData"))
# if (!all.equal(tmp_feats_df, glb_feats_df))
# stop("glb_feats_df r/w not working")
# if (!all.equal(tmp_entity_df, glb_allobs_df))
# stop("glb_allobs_df r/w not working")
rm(split)## Warning in rm(split): object 'split' not foundglb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 9 partition.data.training 6 0 35.356 35.689 0.333
## 10 fit.models 7 0 35.689 NA NAStep 7.0: fit models
# load(paste0(glb_out_pfx, "dsk.RData"))
# keep_cols <- setdiff(names(glb_allobs_df),
# grep("^.src", names(glb_allobs_df), value=TRUE))
# glb_trnobs_df <- glb_allobs_df[glb_allobs_df$.src.trn == 1, keep_cols]
# glb_fitobs_df <- glb_allobs_df[glb_allobs_df$.src.fit == 1, keep_cols]
# glb_OOBobs_df <- glb_allobs_df[glb_allobs_df$.src.OOB == 1, keep_cols]
# glb_newobs_df <- glb_allobs_df[glb_allobs_df$.src.new == 1, keep_cols]
#
# glb_models_lst <- list(); glb_models_df <- data.frame()
#
if (glb_is_classification && glb_is_binomial &&
(length(unique(glb_fitobs_df[, glb_rsp_var])) < 2))
stop("glb_fitobs_df$", glb_rsp_var, ": contains less than 2 unique values: ",
paste0(unique(glb_fitobs_df[, glb_rsp_var]), collapse=", "))
max_cor_y_x_vars <- orderBy(~ -cor.y.abs,
subset(glb_feats_df, (exclude.as.feat == 0) & !is.cor.y.abs.low &
is.na(cor.high.X)))[1:2, "id"]
# while(length(max_cor_y_x_vars) < 2) {
# max_cor_y_x_vars <- c(max_cor_y_x_vars, orderBy(~ -cor.y.abs,
# subset(glb_feats_df, (exclude.as.feat == 0) & !is.cor.y.abs.low))[3, "id"])
# }
if (!is.null(glb_Baseline_mdl_var)) {
if ((max_cor_y_x_vars[1] != glb_Baseline_mdl_var) &
(glb_feats_df[glb_feats_df$id == max_cor_y_x_vars[1], "cor.y.abs"] >
glb_feats_df[glb_feats_df$id == glb_Baseline_mdl_var, "cor.y.abs"]))
stop(max_cor_y_x_vars[1], " has a higher correlation with ", glb_rsp_var,
" than the Baseline var: ", glb_Baseline_mdl_var)
}
glb_model_type <- ifelse(glb_is_regression, "regression", "classification")
# Baseline
if (!is.null(glb_Baseline_mdl_var))
ret_lst <- myfit_mdl(model_id="Baseline",
model_method="mybaseln_classfr",
indep_vars_vctr=glb_Baseline_mdl_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df)
# Most Frequent Outcome "MFO" model: mean(y) for regression
# Not using caret's nullModel since model stats not avl
# Cannot use rpart for multinomial classification since it predicts non-MFO
ret_lst <- myfit_mdl(model_id="MFO",
model_method=ifelse(glb_is_regression, "lm", "myMFO_classfr"),
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df)## [1] "fitting model: MFO.myMFO_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## [1] "in MFO.Classifier$fit"
## [1] "unique.vals:"
## [1] N Y
## Levels: N Y
## [1] "unique.prob:"
## y
## N Y
## 0.8527982 0.1472018
## [1] "MFO.val:"
## [1] "N"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 -none- numeric
## MFO.val 1 -none- character
## x.names 1 -none- character
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] " calling mypredict_mdl for fit:"## Loading required package: ROCR
## Loading required package: gplots
##
## Attaching package: 'gplots'
##
## The following object is masked from 'package:stats':
##
## lowess## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.8527982 0.1472018
## 2 0.8527982 0.1472018
## 3 0.8527982 0.1472018
## 4 0.8527982 0.1472018
## 5 0.8527982 0.1472018
## 6 0.8527982 0.1472018
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.MFO.myMFO_classfr.N
## 1 N 6141
## 2 Y 1060
## Prediction
## Reference N Y
## N 6141 0
## Y 1060 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.527982e-01 0.000000e+00 8.444027e-01 8.609070e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 5.081939e-01 4.439170e-232
## [1] " calling mypredict_mdl for OOB:"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.8527982 0.1472018
## 2 0.8527982 0.1472018
## 3 0.8527982 0.1472018
## 4 0.8527982 0.1472018
## 5 0.8527982 0.1472018
## 6 0.8527982 0.1472018
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.MFO.myMFO_classfr.N
## 1 N 314
## 2 Y 59
## Prediction
## Reference N Y
## N 314 0
## Y 59 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.418231e-01 0.000000e+00 8.007587e-01 8.773551e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 5.346786e-01 4.320720e-14
## model_id model_method feats max.nTuningRuns
## 1 MFO.myMFO_classfr myMFO_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.318 0.003 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8527982
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8444027 0.860907 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8418231
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8007587 0.8773551 0if (glb_is_classification)
# "random" model - only for classification;
# none needed for regression since it is same as MFO
ret_lst <- myfit_mdl(model_id="Random", model_method="myrandom_classfr",
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df)## [1] "fitting model: Random.myrandom_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 table numeric
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] " calling mypredict_mdl for fit:"
## [1] "in Random.Classifier$prob"
## threshold f.score
## 1 0.0 0.2566275
## 2 0.1 0.2566275
## 3 0.2 0.1483932
## 4 0.3 0.1483932
## 5 0.4 0.1483932
## 6 0.5 0.1483932
## 7 0.6 0.1483932
## 8 0.7 0.1483932
## 9 0.8 0.1483932
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Random.myrandom_classfr.Y
## 1 N 6141
## 2 Y 1060
## Prediction
## Reference N Y
## N 0 6141
## Y 0 1060
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.1472018 0.0000000 0.1390930 0.1555973 0.8527982
## AccuracyPValue McnemarPValue
## 1.0000000 0.0000000
## [1] " calling mypredict_mdl for OOB:"
## [1] "in Random.Classifier$prob"
## threshold f.score
## 1 0.0 0.2731481
## 2 0.1 0.2731481
## 3 0.2 0.1864407
## 4 0.3 0.1864407
## 5 0.4 0.1864407
## 6 0.5 0.1864407
## 7 0.6 0.1864407
## 8 0.7 0.1864407
## 9 0.8 0.1864407
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Random.myrandom_classfr.Y
## 1 N 314
## 2 Y 59
## Prediction
## Reference N Y
## N 0 314
## Y 0 59
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 1.581769e-01 0.000000e+00 1.226449e-01 1.992413e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 7.995559e-70
## model_id model_method feats max.nTuningRuns
## 1 Random.myrandom_classfr myrandom_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.235 0.002 0.5008601
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.1 0.2566275 0.1472018
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.139093 0.1555973 0 0.5167872
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.1 0.2731481 0.1581769
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.1226449 0.1992413 0# Any models that have tuning parameters has "better" results with cross-validation
# (except rf) & "different" results for different outcome metrics
# Max.cor.Y
# Check impact of cv
# rpart is not a good candidate since caret does not optimize cp (only tuning parameter of rpart) well
ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_vars,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df)## [1] "fitting model: Max.cor.Y.cv.0.rpart"
## [1] " indep_vars: timbre_6_min, timbre_4_max"## Loading required package: rpart## Fitting cp = 0.00108 on full training set## Loading required package: rpart.plot
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 7201
##
## CP nsplit rel error
## 1 0.001078167 0 1
##
## Node number 1: 7201 observations
## predicted class=N expected loss=0.1472018 P(node) =1
## class counts: 6141 1060
## probabilities: 0.853 0.147
##
## n= 7201
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 7201 1060 N (0.8527982 0.1472018) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 6141
## 2 Y 1060
## Prediction
## Reference N Y
## N 6141 0
## Y 1060 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.527982e-01 0.000000e+00 8.444027e-01 8.609070e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 5.081939e-01 4.439170e-232
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 314
## 2 Y 59
## Prediction
## Reference N Y
## N 314 0
## Y 59 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.418231e-01 0.000000e+00 8.007587e-01 8.773551e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 5.346786e-01 4.320720e-14
## model_id model_method feats
## 1 Max.cor.Y.cv.0.rpart rpart timbre_6_min, timbre_4_max
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.843 0.124
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0.5 0 0.8527982
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8444027 0.860907 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8418231
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8007587 0.8773551 0ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0.cp.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_vars,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=0,
tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))## [1] "fitting model: Max.cor.Y.cv.0.cp.0.rpart"
## [1] " indep_vars: timbre_6_min, timbre_4_max"
## Fitting cp = 0 on full training set## Warning: labs do not fit even at cex 0.15, there may be some overplotting
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 7201
##
## CP nsplit rel error
## 1 0.0010781671 0 1.0000000
## 2 0.0009433962 35 0.9509434
## 3 0.0008086253 50 0.9358491
## 4 0.0006738544 59 0.9283019
## 5 0.0006289308 74 0.9103774
## 6 0.0005390836 78 0.9075472
## 7 0.0004716981 99 0.8924528
## 8 0.0003144654 137 0.8650943
## 9 0.0002358491 143 0.8632075
## 10 0.0001347709 155 0.8603774
## 11 0.0000000000 162 0.8594340
##
## Variable importance
## timbre_6_min timbre_4_max
## 55 45
##
## Node number 1: 7201 observations, complexity param=0.001078167
## predicted class=N expected loss=0.1472018 P(node) =1
## class counts: 6141 1060
## probabilities: 0.853 0.147
## left son=2 (4382 obs) right son=3 (2819 obs)
## Primary splits:
## timbre_6_min < -86.184 to the right, improve=63.85999, (0 missing)
## timbre_4_max < 90.6385 to the left, improve=57.81832, (0 missing)
## Surrogate splits:
## timbre_4_max < 111.935 to the left, agree=0.68, adj=0.182, (0 split)
##
## Node number 2: 4382 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.09379279 P(node) =0.6085266
## class counts: 3971 411
## probabilities: 0.906 0.094
## left son=4 (1731 obs) right son=5 (2651 obs)
## Primary splits:
## timbre_4_max < 87.318 to the left, improve=17.36542, (0 missing)
## timbre_6_min < -64.667 to the right, improve=13.80606, (0 missing)
## Surrogate splits:
## timbre_6_min < -60.748 to the right, agree=0.658, adj=0.135, (0 split)
##
## Node number 3: 2819 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2302235 P(node) =0.3914734
## class counts: 2170 649
## probabilities: 0.770 0.230
## left son=6 (898 obs) right son=7 (1921 obs)
## Primary splits:
## timbre_4_max < 109.353 to the left, improve=18.257260, (0 missing)
## timbre_6_min < -110.867 to the right, improve= 9.658228, (0 missing)
## Surrogate splits:
## timbre_6_min < -86.831 to the right, agree=0.683, adj=0.003, (0 split)
##
## Node number 4: 1731 observations
## predicted class=N expected loss=0.03870595 P(node) =0.2403833
## class counts: 1664 67
## probabilities: 0.961 0.039
##
## Node number 5: 2651 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1297624 P(node) =0.3681433
## class counts: 2307 344
## probabilities: 0.870 0.130
## left son=10 (701 obs) right son=11 (1950 obs)
## Primary splits:
## timbre_6_min < -64.667 to the right, improve=8.922924, (0 missing)
## timbre_4_max < 189.9985 to the left, improve=1.376515, (0 missing)
##
## Node number 6: 898 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.1469933 P(node) =0.1247049
## class counts: 766 132
## probabilities: 0.853 0.147
## left son=12 (366 obs) right son=13 (532 obs)
## Primary splits:
## timbre_4_max < 91.4875 to the left, improve=3.259933, (0 missing)
## timbre_6_min < -96.5325 to the right, improve=2.574484, (0 missing)
## Surrogate splits:
## timbre_6_min < -86.4725 to the right, agree=0.598, adj=0.014, (0 split)
##
## Node number 7: 1921 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2691307 P(node) =0.2667685
## class counts: 1404 517
## probabilities: 0.731 0.269
## left son=14 (1522 obs) right son=15 (399 obs)
## Primary splits:
## timbre_6_min < -112.424 to the right, improve=5.549425, (0 missing)
## timbre_4_max < 135.6385 to the left, improve=2.789776, (0 missing)
## Surrogate splits:
## timbre_4_max < 109.422 to the right, agree=0.793, adj=0.003, (0 split)
##
## Node number 10: 701 observations
## predicted class=N expected loss=0.06134094 P(node) =0.09734759
## class counts: 658 43
## probabilities: 0.939 0.061
##
## Node number 11: 1950 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.154359 P(node) =0.2707957
## class counts: 1649 301
## probabilities: 0.846 0.154
## left son=22 (1928 obs) right son=23 (22 obs)
## Primary splits:
## timbre_4_max < 189.9985 to the left, improve=1.949059, (0 missing)
## timbre_6_min < -85.6325 to the left, improve=1.388632, (0 missing)
##
## Node number 12: 366 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.09562842 P(node) =0.05082627
## class counts: 331 35
## probabilities: 0.904 0.096
## left son=24 (92 obs) right son=25 (274 obs)
## Primary splits:
## timbre_4_max < 67.6285 to the left, improve=0.9761157, (0 missing)
## timbre_6_min < -86.478 to the left, improve=0.7113886, (0 missing)
## Surrogate splits:
## timbre_6_min < -119.2955 to the left, agree=0.754, adj=0.022, (0 split)
##
## Node number 13: 532 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.1823308 P(node) =0.07387863
## class counts: 435 97
## probabilities: 0.818 0.182
## left son=26 (522 obs) right son=27 (10 obs)
## Primary splits:
## timbre_4_max < 91.831 to the right, improve=3.555789, (0 missing)
## timbre_6_min < -110.0435 to the right, improve=2.459997, (0 missing)
##
## Node number 14: 1522 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2496715 P(node) =0.2113595
## class counts: 1142 380
## probabilities: 0.750 0.250
## left son=28 (1514 obs) right son=29 (8 obs)
## Primary splits:
## timbre_4_max < 109.674 to the right, improve=2.265854, (0 missing)
## timbre_6_min < -90.2195 to the left, improve=1.591299, (0 missing)
##
## Node number 15: 399 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3433584 P(node) =0.05540897
## class counts: 262 137
## probabilities: 0.657 0.343
## left son=30 (385 obs) right son=31 (14 obs)
## Primary splits:
## timbre_6_min < -112.758 to the left, improve=1.50941, (0 missing)
## timbre_4_max < 124.0415 to the left, improve=1.36934, (0 missing)
##
## Node number 22: 1928 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.151971 P(node) =0.2677406
## class counts: 1635 293
## probabilities: 0.848 0.152
## left son=44 (51 obs) right son=45 (1877 obs)
## Primary splits:
## timbre_6_min < -85.6325 to the left, improve=1.332038, (0 missing)
## timbre_4_max < 87.524 to the right, improve=1.075013, (0 missing)
##
## Node number 23: 22 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.3636364 P(node) =0.003055131
## class counts: 14 8
## probabilities: 0.636 0.364
## left son=46 (7 obs) right son=47 (15 obs)
## Primary splits:
## timbre_4_max < 204.9105 to the right, improve=2.715152, (0 missing)
## timbre_6_min < -76.618 to the right, improve=2.548485, (0 missing)
##
## Node number 24: 92 observations
## predicted class=N expected loss=0.0326087 P(node) =0.012776
## class counts: 89 3
## probabilities: 0.967 0.033
##
## Node number 25: 274 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.1167883 P(node) =0.03805027
## class counts: 242 32
## probabilities: 0.883 0.117
## left son=50 (253 obs) right son=51 (21 obs)
## Primary splits:
## timbre_4_max < 71.4975 to the right, improve=2.132926, (0 missing)
## timbre_6_min < -86.704 to the left, improve=1.396616, (0 missing)
##
## Node number 26: 522 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.1743295 P(node) =0.07248993
## class counts: 431 91
## probabilities: 0.826 0.174
## left son=52 (434 obs) right son=53 (88 obs)
## Primary splits:
## timbre_6_min < -110.0435 to the right, improve=2.049576, (0 missing)
## timbre_4_max < 108.1615 to the right, improve=1.330437, (0 missing)
##
## Node number 27: 10 observations
## predicted class=Y expected loss=0.4 P(node) =0.001388696
## class counts: 4 6
## probabilities: 0.400 0.600
##
## Node number 28: 1514 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2476882 P(node) =0.2102486
## class counts: 1139 375
## probabilities: 0.752 0.248
## left son=56 (895 obs) right son=57 (619 obs)
## Primary splits:
## timbre_4_max < 139.04 to the left, improve=2.337676, (0 missing)
## timbre_6_min < -90.081 to the left, improve=1.600316, (0 missing)
## Surrogate splits:
## timbre_6_min < -110.69 to the right, agree=0.598, adj=0.018, (0 split)
##
## Node number 29: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 30: 385 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3350649 P(node) =0.0534648
## class counts: 256 129
## probabilities: 0.665 0.335
## left son=60 (179 obs) right son=61 (206 obs)
## Primary splits:
## timbre_4_max < 135.5785 to the left, improve=1.6826600, (0 missing)
## timbre_6_min < -121.5495 to the right, improve=0.6027706, (0 missing)
## Surrogate splits:
## timbre_6_min < -117.412 to the right, agree=0.595, adj=0.128, (0 split)
##
## Node number 31: 14 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.001944174
## class counts: 6 8
## probabilities: 0.429 0.571
##
## Node number 44: 51 observations
## predicted class=N expected loss=0.03921569 P(node) =0.00708235
## class counts: 49 2
## probabilities: 0.961 0.039
##
## Node number 45: 1877 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1550346 P(node) =0.2606582
## class counts: 1586 291
## probabilities: 0.845 0.155
## left son=90 (1847 obs) right son=91 (30 obs)
## Primary splits:
## timbre_6_min < -85.353 to the right, improve=3.658963, (0 missing)
## timbre_4_max < 87.524 to the right, improve=1.051434, (0 missing)
##
## Node number 46: 7 observations
## predicted class=N expected loss=0 P(node) =0.0009720872
## class counts: 7 0
## probabilities: 1.000 0.000
##
## Node number 47: 15 observations
## predicted class=Y expected loss=0.4666667 P(node) =0.002083044
## class counts: 7 8
## probabilities: 0.467 0.533
##
## Node number 50: 253 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.09881423 P(node) =0.03513401
## class counts: 228 25
## probabilities: 0.901 0.099
## left son=100 (242 obs) right son=101 (11 obs)
## Primary splits:
## timbre_6_min < -86.704 to the left, improve=1.6130080, (0 missing)
## timbre_4_max < 90.1555 to the right, improve=0.6646697, (0 missing)
##
## Node number 51: 21 observations
## predicted class=N expected loss=0.3333333 P(node) =0.002916262
## class counts: 14 7
## probabilities: 0.667 0.333
##
## Node number 52: 434 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.1543779 P(node) =0.06026941
## class counts: 367 67
## probabilities: 0.846 0.154
## left son=104 (98 obs) right son=105 (336 obs)
## Primary splits:
## timbre_4_max < 105.5845 to the right, improve=1.7419350, (0 missing)
## timbre_6_min < -96.5325 to the right, improve=0.6145578, (0 missing)
##
## Node number 53: 88 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.2727273 P(node) =0.01222052
## class counts: 64 24
## probabilities: 0.727 0.273
## left son=106 (9 obs) right son=107 (79 obs)
## Primary splits:
## timbre_4_max < 107.9985 to the right, improve=1.4913690, (0 missing)
## timbre_6_min < -110.881 to the left, improve=0.9090909, (0 missing)
##
## Node number 56: 895 observations, complexity param=0.001078167
## predicted class=N expected loss=0.224581 P(node) =0.1242883
## class counts: 694 201
## probabilities: 0.775 0.225
## left son=112 (31 obs) right son=113 (864 obs)
## Primary splits:
## timbre_4_max < 137.6295 to the right, improve=1.6454820, (0 missing)
## timbre_6_min < -110.683 to the left, improve=0.8060126, (0 missing)
##
## Node number 57: 619 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2810985 P(node) =0.08596028
## class counts: 445 174
## probabilities: 0.719 0.281
## left son=114 (525 obs) right son=115 (94 obs)
## Primary splits:
## timbre_6_min < -90.091 to the left, improve=2.806318, (0 missing)
## timbre_4_max < 224.1235 to the right, improve=1.280816, (0 missing)
##
## Node number 60: 179 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2849162 P(node) =0.02485766
## class counts: 128 51
## probabilities: 0.715 0.285
## left son=120 (9 obs) right son=121 (170 obs)
## Primary splits:
## timbre_4_max < 134.8965 to the right, improve=1.538547, (0 missing)
## timbre_6_min < -117.385 to the left, improve=1.167730, (0 missing)
##
## Node number 61: 206 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3786408 P(node) =0.02860714
## class counts: 128 78
## probabilities: 0.621 0.379
## left son=122 (198 obs) right son=123 (8 obs)
## Primary splits:
## timbre_4_max < 137.1985 to the right, improve=2.295675, (0 missing)
## timbre_6_min < -120.7485 to the right, improve=1.017479, (0 missing)
##
## Node number 90: 1847 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1510558 P(node) =0.2564922
## class counts: 1568 279
## probabilities: 0.849 0.151
## left son=180 (1175 obs) right son=181 (672 obs)
## Primary splits:
## timbre_6_min < -78.8265 to the right, improve=0.7298795, (0 missing)
## timbre_4_max < 141.9505 to the left, improve=0.4721727, (0 missing)
## Surrogate splits:
## timbre_4_max < 87.4895 to the right, agree=0.638, adj=0.004, (0 split)
##
## Node number 91: 30 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.4 P(node) =0.004166088
## class counts: 18 12
## probabilities: 0.600 0.400
## left son=182 (19 obs) right son=183 (11 obs)
## Primary splits:
## timbre_6_min < -85.494 to the left, improve=1.94067, (0 missing)
## timbre_4_max < 128.539 to the right, improve=0.90000, (0 missing)
## Surrogate splits:
## timbre_4_max < 126.7505 to the right, agree=0.667, adj=0.091, (0 split)
##
## Node number 100: 242 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.08677686 P(node) =0.03360644
## class counts: 221 21
## probabilities: 0.913 0.087
## left son=200 (138 obs) right son=201 (104 obs)
## Primary splits:
## timbre_6_min < -96.956 to the right, improve=0.8347476, (0 missing)
## timbre_4_max < 90.1555 to the right, improve=0.5157493, (0 missing)
## Surrogate splits:
## timbre_4_max < 90.576 to the left, agree=0.595, adj=0.058, (0 split)
##
## Node number 101: 11 observations
## predicted class=N expected loss=0.3636364 P(node) =0.001527566
## class counts: 7 4
## probabilities: 0.636 0.364
##
## Node number 104: 98 observations
## predicted class=N expected loss=0.07142857 P(node) =0.01360922
## class counts: 91 7
## probabilities: 0.929 0.071
##
## Node number 105: 336 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.1785714 P(node) =0.04666019
## class counts: 276 60
## probabilities: 0.821 0.179
## left son=210 (250 obs) right son=211 (86 obs)
## Primary splits:
## timbre_4_max < 102.463 to the left, improve=0.9952425, (0 missing)
## timbre_6_min < -95.538 to the right, improve=0.7739358, (0 missing)
## Surrogate splits:
## timbre_6_min < -109.6235 to the right, agree=0.747, adj=0.012, (0 split)
##
## Node number 106: 9 observations
## predicted class=N expected loss=0 P(node) =0.001249826
## class counts: 9 0
## probabilities: 1.000 0.000
##
## Node number 107: 79 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.3037975 P(node) =0.0109707
## class counts: 55 24
## probabilities: 0.696 0.304
## left son=214 (53 obs) right son=215 (26 obs)
## Primary splits:
## timbre_4_max < 104.347 to the left, improve=2.983759, (0 missing)
## timbre_6_min < -110.881 to the left, improve=1.100261, (0 missing)
## Surrogate splits:
## timbre_6_min < -110.6185 to the left, agree=0.684, adj=0.038, (0 split)
##
## Node number 112: 31 observations
## predicted class=N expected loss=0.06451613 P(node) =0.004304958
## class counts: 29 2
## probabilities: 0.935 0.065
##
## Node number 113: 864 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2303241 P(node) =0.1199833
## class counts: 665 199
## probabilities: 0.770 0.230
## left son=226 (371 obs) right son=227 (493 obs)
## Primary splits:
## timbre_4_max < 121.5665 to the left, improve=1.0317510, (0 missing)
## timbre_6_min < -102.7925 to the right, improve=0.6989913, (0 missing)
## Surrogate splits:
## timbre_6_min < -110.7055 to the left, agree=0.574, adj=0.008, (0 split)
##
## Node number 114: 525 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2609524 P(node) =0.07290654
## class counts: 388 137
## probabilities: 0.739 0.261
## left son=228 (25 obs) right son=229 (500 obs)
## Primary splits:
## timbre_6_min < -91.2335 to the right, improve=2.563048, (0 missing)
## timbre_4_max < 224.1235 to the right, improve=1.106398, (0 missing)
##
## Node number 115: 94 observations, complexity param=0.001078167
## predicted class=N expected loss=0.393617 P(node) =0.01305374
## class counts: 57 37
## probabilities: 0.606 0.394
## left son=230 (62 obs) right son=231 (32 obs)
## Primary splits:
## timbre_6_min < -89.0365 to the right, improve=2.767502, (0 missing)
## timbre_4_max < 141.9595 to the right, improve=2.221178, (0 missing)
##
## Node number 120: 9 observations
## predicted class=N expected loss=0 P(node) =0.001249826
## class counts: 9 0
## probabilities: 1.000 0.000
##
## Node number 121: 170 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3 P(node) =0.02360783
## class counts: 119 51
## probabilities: 0.700 0.300
## left son=242 (97 obs) right son=243 (73 obs)
## Primary splits:
## timbre_6_min < -117.385 to the left, improve=1.2488910, (0 missing)
## timbre_4_max < 124.0415 to the left, improve=0.6783251, (0 missing)
## Surrogate splits:
## timbre_4_max < 110.5345 to the right, agree=0.594, adj=0.055, (0 split)
##
## Node number 122: 198 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3636364 P(node) =0.02749618
## class counts: 126 72
## probabilities: 0.636 0.364
## left son=244 (23 obs) right son=245 (175 obs)
## Primary splits:
## timbre_6_min < -134.7425 to the left, improve=1.1131340, (0 missing)
## timbre_4_max < 180.7105 to the right, improve=0.7053292, (0 missing)
##
## Node number 123: 8 observations
## predicted class=Y expected loss=0.25 P(node) =0.001110957
## class counts: 2 6
## probabilities: 0.250 0.750
##
## Node number 180: 1175 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1404255 P(node) =0.1631718
## class counts: 1010 165
## probabilities: 0.860 0.140
## left son=360 (1118 obs) right son=361 (57 obs)
## Primary splits:
## timbre_6_min < -65.368 to the left, improve=0.9203472, (0 missing)
## timbre_4_max < 119.0535 to the left, improve=0.5208527, (0 missing)
##
## Node number 181: 672 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1696429 P(node) =0.09332037
## class counts: 558 114
## probabilities: 0.830 0.170
## left son=362 (614 obs) right son=363 (58 obs)
## Primary splits:
## timbre_6_min < -79.48 to the left, improve=1.9351540, (0 missing)
## timbre_4_max < 93.8785 to the left, improve=0.5562932, (0 missing)
##
## Node number 182: 19 observations
## predicted class=N expected loss=0.2631579 P(node) =0.002638522
## class counts: 14 5
## probabilities: 0.737 0.263
##
## Node number 183: 11 observations
## predicted class=Y expected loss=0.3636364 P(node) =0.001527566
## class counts: 4 7
## probabilities: 0.364 0.636
##
## Node number 200: 138 observations
## predicted class=N expected loss=0.05072464 P(node) =0.019164
## class counts: 131 7
## probabilities: 0.949 0.051
##
## Node number 201: 104 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.1346154 P(node) =0.01444244
## class counts: 90 14
## probabilities: 0.865 0.135
## left son=402 (27 obs) right son=403 (77 obs)
## Primary splits:
## timbre_4_max < 88.1385 to the right, improve=1.3216780, (0 missing)
## timbre_6_min < -103.526 to the left, improve=0.4986264, (0 missing)
##
## Node number 210: 250 observations
## predicted class=N expected loss=0.156 P(node) =0.0347174
## class counts: 211 39
## probabilities: 0.844 0.156
##
## Node number 211: 86 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.244186 P(node) =0.01194279
## class counts: 65 21
## probabilities: 0.756 0.244
## left son=422 (26 obs) right son=423 (60 obs)
## Primary splits:
## timbre_6_min < -98.182 to the left, improve=2.0852120, (0 missing)
## timbre_4_max < 102.639 to the right, improve=0.5181463, (0 missing)
##
## Node number 214: 53 observations
## predicted class=N expected loss=0.2075472 P(node) =0.007360089
## class counts: 42 11
## probabilities: 0.792 0.208
##
## Node number 215: 26 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.5 P(node) =0.00361061
## class counts: 13 13
## probabilities: 0.500 0.500
## left son=430 (9 obs) right son=431 (17 obs)
## Primary splits:
## timbre_6_min < -116.569 to the left, improve=2.1241830, (0 missing)
## timbre_4_max < 107.254 to the left, improve=0.3611111, (0 missing)
## Surrogate splits:
## timbre_4_max < 105.1215 to the left, agree=0.769, adj=0.333, (0 split)
##
## Node number 226: 371 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.2021563 P(node) =0.05152062
## class counts: 296 75
## probabilities: 0.798 0.202
## left son=452 (129 obs) right son=453 (242 obs)
## Primary splits:
## timbre_4_max < 117.926 to the right, improve=1.9588230, (0 missing)
## timbre_6_min < -110.6995 to the left, improve=0.9265499, (0 missing)
##
## Node number 227: 493 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2515213 P(node) =0.06846271
## class counts: 369 124
## probabilities: 0.748 0.252
## left son=454 (391 obs) right son=455 (102 obs)
## Primary splits:
## timbre_4_max < 125.184 to the right, improve=1.3337150, (0 missing)
## timbre_6_min < -89.5705 to the left, improve=0.7310924, (0 missing)
##
## Node number 228: 25 observations
## predicted class=N expected loss=0.04 P(node) =0.00347174
## class counts: 24 1
## probabilities: 0.960 0.040
##
## Node number 229: 500 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.272 P(node) =0.0694348
## class counts: 364 136
## probabilities: 0.728 0.272
## left son=458 (8 obs) right son=459 (492 obs)
## Primary splits:
## timbre_6_min < -111.831 to the left, improve=1.202992, (0 missing)
## timbre_4_max < 224.1235 to the right, improve=1.202992, (0 missing)
##
## Node number 230: 62 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3064516 P(node) =0.008609915
## class counts: 43 19
## probabilities: 0.694 0.306
## left son=460 (55 obs) right son=461 (7 obs)
## Primary splits:
## timbre_4_max < 142.605 to the right, improve=2.624969, (0 missing)
## timbre_6_min < -88.4605 to the left, improve=2.275406, (0 missing)
##
## Node number 231: 32 observations, complexity param=0.001078167
## predicted class=Y expected loss=0.4375 P(node) =0.004443827
## class counts: 14 18
## probabilities: 0.438 0.562
## left son=462 (20 obs) right son=463 (12 obs)
## Primary splits:
## timbre_4_max < 162.371 to the left, improve=1.3500000, (0 missing)
## timbre_6_min < -89.69 to the left, improve=0.3906926, (0 missing)
## Surrogate splits:
## timbre_6_min < -90.062 to the right, agree=0.688, adj=0.167, (0 split)
##
## Node number 242: 97 observations, complexity param=0.001078167
## predicted class=N expected loss=0.2474227 P(node) =0.01347035
## class counts: 73 24
## probabilities: 0.753 0.247
## left son=484 (75 obs) right son=485 (22 obs)
## Primary splits:
## timbre_6_min < -130.7685 to the right, improve=1.487348, (0 missing)
## timbre_4_max < 124.1495 to the left, improve=1.235154, (0 missing)
##
## Node number 243: 73 observations, complexity param=0.001078167
## predicted class=N expected loss=0.369863 P(node) =0.01013748
## class counts: 46 27
## probabilities: 0.630 0.370
## left son=486 (25 obs) right son=487 (48 obs)
## Primary splits:
## timbre_6_min < -114.1775 to the right, improve=1.2823970, (0 missing)
## timbre_4_max < 113.2695 to the left, improve=0.4126432, (0 missing)
## Surrogate splits:
## timbre_4_max < 110.273 to the left, agree=0.699, adj=0.12, (0 split)
##
## Node number 244: 23 observations
## predicted class=N expected loss=0.2173913 P(node) =0.003194001
## class counts: 18 5
## probabilities: 0.783 0.217
##
## Node number 245: 175 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3828571 P(node) =0.02430218
## class counts: 108 67
## probabilities: 0.617 0.383
## left son=490 (87 obs) right son=491 (88 obs)
## Primary splits:
## timbre_6_min < -120.7485 to the right, improve=1.8194000, (0 missing)
## timbre_4_max < 169.243 to the left, improve=0.8804762, (0 missing)
## Surrogate splits:
## timbre_4_max < 156.5615 to the right, agree=0.577, adj=0.149, (0 split)
##
## Node number 360: 1118 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.1359571 P(node) =0.1552562
## class counts: 966 152
## probabilities: 0.864 0.136
## left son=720 (38 obs) right son=721 (1080 obs)
## Primary splits:
## timbre_6_min < -65.843 to the right, improve=0.9457575, (0 missing)
## timbre_4_max < 119.0535 to the left, improve=0.6736932, (0 missing)
##
## Node number 361: 57 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2280702 P(node) =0.007915567
## class counts: 44 13
## probabilities: 0.772 0.228
## left son=722 (45 obs) right son=723 (12 obs)
## Primary splits:
## timbre_6_min < -65.235 to the right, improve=3.8368420, (0 missing)
## timbre_4_max < 110.9155 to the right, improve=0.7368421, (0 missing)
##
## Node number 362: 614 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1579805 P(node) =0.08526594
## class counts: 517 97
## probabilities: 0.842 0.158
## left son=724 (551 obs) right son=725 (63 obs)
## Primary splits:
## timbre_4_max < 95.1505 to the right, improve=0.5794584, (0 missing)
## timbre_6_min < -81.6945 to the left, improve=0.5584624, (0 missing)
##
## Node number 363: 58 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2931034 P(node) =0.008054437
## class counts: 41 17
## probabilities: 0.707 0.293
## left son=726 (45 obs) right son=727 (13 obs)
## Primary splits:
## timbre_6_min < -79.4275 to the right, improve=2.017389, (0 missing)
## timbre_4_max < 94.066 to the left, improve=1.367816, (0 missing)
##
## Node number 402: 27 observations
## predicted class=N expected loss=0 P(node) =0.003749479
## class counts: 27 0
## probabilities: 1.000 0.000
##
## Node number 403: 77 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.1818182 P(node) =0.01069296
## class counts: 63 14
## probabilities: 0.818 0.182
## left son=806 (52 obs) right son=807 (25 obs)
## Primary splits:
## timbre_4_max < 84.6245 to the left, improve=2.3506290, (0 missing)
## timbre_6_min < -103.534 to the left, improve=0.5374693, (0 missing)
## Surrogate splits:
## timbre_6_min < -98.749 to the left, agree=0.688, adj=0.04, (0 split)
##
## Node number 422: 26 observations
## predicted class=N expected loss=0.07692308 P(node) =0.00361061
## class counts: 24 2
## probabilities: 0.923 0.077
##
## Node number 423: 60 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.3166667 P(node) =0.008332176
## class counts: 41 19
## probabilities: 0.683 0.317
## left son=846 (52 obs) right son=847 (8 obs)
## Primary splits:
## timbre_6_min < -96.3905 to the right, improve=3.4666670, (0 missing)
## timbre_4_max < 102.7715 to the right, improve=0.6205128, (0 missing)
##
## Node number 430: 9 observations
## predicted class=N expected loss=0.2222222 P(node) =0.001249826
## class counts: 7 2
## probabilities: 0.778 0.222
##
## Node number 431: 17 observations
## predicted class=Y expected loss=0.3529412 P(node) =0.002360783
## class counts: 6 11
## probabilities: 0.353 0.647
##
## Node number 452: 129 observations
## predicted class=N expected loss=0.1317829 P(node) =0.01791418
## class counts: 112 17
## probabilities: 0.868 0.132
##
## Node number 453: 242 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.2396694 P(node) =0.03360644
## class counts: 184 58
## probabilities: 0.760 0.240
## left son=906 (217 obs) right son=907 (25 obs)
## Primary splits:
## timbre_4_max < 116.948 to the left, improve=1.4333700, (0 missing)
## timbre_6_min < -110.6995 to the left, improve=0.8281343, (0 missing)
##
## Node number 454: 391 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2327366 P(node) =0.05429801
## class counts: 300 91
## probabilities: 0.767 0.233
## left son=908 (18 obs) right son=909 (373 obs)
## Primary splits:
## timbre_4_max < 125.5545 to the left, improve=1.1846900, (0 missing)
## timbre_6_min < -102.915 to the right, improve=0.8799361, (0 missing)
##
## Node number 455: 102 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3235294 P(node) =0.0141647
## class counts: 69 33
## probabilities: 0.676 0.324
## left son=910 (19 obs) right son=911 (83 obs)
## Primary splits:
## timbre_6_min < -104.638 to the left, improve=1.281174, (0 missing)
## timbre_4_max < 124.123 to the left, improve=1.024837, (0 missing)
##
## Node number 458: 8 observations
## predicted class=N expected loss=0 P(node) =0.001110957
## class counts: 8 0
## probabilities: 1.000 0.000
##
## Node number 459: 492 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2764228 P(node) =0.06832384
## class counts: 356 136
## probabilities: 0.724 0.276
## left son=918 (473 obs) right son=919 (19 obs)
## Primary splits:
## timbre_6_min < -110.8735 to the right, improve=2.468288, (0 missing)
## timbre_4_max < 224.1235 to the right, improve=1.242760, (0 missing)
##
## Node number 460: 55 observations, complexity param=0.0006289308
## predicted class=N expected loss=0.2545455 P(node) =0.007637828
## class counts: 41 14
## probabilities: 0.745 0.255
## left son=920 (14 obs) right son=921 (41 obs)
## Primary splits:
## timbre_6_min < -88.4635 to the left, improve=1.2594870, (0 missing)
## timbre_4_max < 147.1105 to the left, improve=0.7363636, (0 missing)
##
## Node number 461: 7 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.0009720872
## class counts: 2 5
## probabilities: 0.286 0.714
##
## Node number 462: 20 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.45 P(node) =0.002777392
## class counts: 11 9
## probabilities: 0.550 0.450
## left son=924 (7 obs) right son=925 (13 obs)
## Primary splits:
## timbre_4_max < 147.704 to the right, improve=0.5813187, (0 missing)
## timbre_6_min < -89.461 to the left, improve=0.3646465, (0 missing)
##
## Node number 463: 12 observations
## predicted class=Y expected loss=0.25 P(node) =0.001666435
## class counts: 3 9
## probabilities: 0.250 0.750
##
## Node number 484: 75 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2 P(node) =0.01041522
## class counts: 60 15
## probabilities: 0.800 0.200
## left son=968 (17 obs) right son=969 (58 obs)
## Primary splits:
## timbre_6_min < -119.76 to the right, improve=0.8762677, (0 missing)
## timbre_4_max < 117.565 to the right, improve=0.8696398, (0 missing)
##
## Node number 485: 22 observations, complexity param=0.001078167
## predicted class=N expected loss=0.4090909 P(node) =0.003055131
## class counts: 13 9
## probabilities: 0.591 0.409
## left son=970 (15 obs) right son=971 (7 obs)
## Primary splits:
## timbre_4_max < 126.721 to the left, improve=1.9125540, (0 missing)
## timbre_6_min < -135.49 to the left, improve=0.6534577, (0 missing)
## Surrogate splits:
## timbre_6_min < -131.8545 to the left, agree=0.773, adj=0.286, (0 split)
##
## Node number 486: 25 observations
## predicted class=N expected loss=0.24 P(node) =0.00347174
## class counts: 19 6
## probabilities: 0.760 0.240
##
## Node number 487: 48 observations, complexity param=0.001078167
## predicted class=N expected loss=0.4375 P(node) =0.006665741
## class counts: 27 21
## probabilities: 0.562 0.438
## left son=974 (28 obs) right son=975 (20 obs)
## Primary splits:
## timbre_4_max < 121.1635 to the right, improve=1.8107140, (0 missing)
## timbre_6_min < -116.3835 to the left, improve=0.8101852, (0 missing)
## Surrogate splits:
## timbre_6_min < -117.071 to the right, agree=0.646, adj=0.15, (0 split)
##
## Node number 490: 87 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3103448 P(node) =0.01208166
## class counts: 60 27
## probabilities: 0.690 0.310
## left son=980 (65 obs) right son=981 (22 obs)
## Primary splits:
## timbre_4_max < 167.48 to the left, improve=1.2245960, (0 missing)
## timbre_6_min < -119.501 to the left, improve=0.7969349, (0 missing)
## Surrogate splits:
## timbre_6_min < -120.65 to the right, agree=0.759, adj=0.045, (0 split)
##
## Node number 491: 88 observations, complexity param=0.001078167
## predicted class=N expected loss=0.4545455 P(node) =0.01222052
## class counts: 48 40
## probabilities: 0.545 0.455
## left son=982 (53 obs) right son=983 (35 obs)
## Primary splits:
## timbre_6_min < -125.0005 to the left, improve=0.9064445, (0 missing)
## timbre_4_max < 145.766 to the right, improve=0.7627850, (0 missing)
## Surrogate splits:
## timbre_4_max < 140.308 to the right, agree=0.682, adj=0.2, (0 split)
##
## Node number 720: 38 observations
## predicted class=N expected loss=0.02631579 P(node) =0.005277045
## class counts: 37 1
## probabilities: 0.974 0.026
##
## Node number 721: 1080 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.1398148 P(node) =0.1499792
## class counts: 929 151
## probabilities: 0.860 0.140
## left son=1442 (885 obs) right son=1443 (195 obs)
## Primary splits:
## timbre_6_min < -68.5755 to the left, improve=0.9552393, (0 missing)
## timbre_4_max < 119.0535 to the left, improve=0.6692646, (0 missing)
##
## Node number 722: 45 observations
## predicted class=N expected loss=0.1333333 P(node) =0.006249132
## class counts: 39 6
## probabilities: 0.867 0.133
##
## Node number 723: 12 observations
## predicted class=Y expected loss=0.4166667 P(node) =0.001666435
## class counts: 5 7
## probabilities: 0.417 0.583
##
## Node number 724: 551 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1506352 P(node) =0.07651715
## class counts: 468 83
## probabilities: 0.849 0.151
## left son=1448 (340 obs) right son=1449 (211 obs)
## Primary splits:
## timbre_6_min < -81.6945 to the left, improve=0.7998523, (0 missing)
## timbre_4_max < 156.9115 to the left, improve=0.4770407, (0 missing)
## Surrogate splits:
## timbre_4_max < 95.713 to the right, agree=0.619, adj=0.005, (0 split)
##
## Node number 725: 63 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2222222 P(node) =0.008748785
## class counts: 49 14
## probabilities: 0.778 0.222
## left son=1450 (51 obs) right son=1451 (12 obs)
## Primary splits:
## timbre_4_max < 93.8785 to the left, improve=5.856209, (0 missing)
## timbre_6_min < -83.969 to the right, improve=1.162945, (0 missing)
##
## Node number 726: 45 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2222222 P(node) =0.006249132
## class counts: 35 10
## probabilities: 0.778 0.222
## left son=1452 (24 obs) right son=1453 (21 obs)
## Primary splits:
## timbre_6_min < -79.1945 to the left, improve=0.9722222, (0 missing)
## timbre_4_max < 132.956 to the right, improve=0.5020796, (0 missing)
## Surrogate splits:
## timbre_4_max < 129.0205 to the left, agree=0.756, adj=0.476, (0 split)
##
## Node number 727: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.001805305
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 806: 52 observations
## predicted class=N expected loss=0.09615385 P(node) =0.007221219
## class counts: 47 5
## probabilities: 0.904 0.096
##
## Node number 807: 25 observations, complexity param=0.0001347709
## predicted class=N expected loss=0.36 P(node) =0.00347174
## class counts: 16 9
## probabilities: 0.640 0.360
## left son=1614 (12 obs) right son=1615 (13 obs)
## Primary splits:
## timbre_6_min < -103.9 to the left, improve=1.7251280, (0 missing)
## timbre_4_max < 86.511 to the left, improve=0.4611765, (0 missing)
## Surrogate splits:
## timbre_4_max < 85.8535 to the left, agree=0.64, adj=0.25, (0 split)
##
## Node number 846: 52 observations
## predicted class=N expected loss=0.25 P(node) =0.007221219
## class counts: 39 13
## probabilities: 0.750 0.250
##
## Node number 847: 8 observations
## predicted class=Y expected loss=0.25 P(node) =0.001110957
## class counts: 2 6
## probabilities: 0.250 0.750
##
## Node number 906: 217 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.2211982 P(node) =0.0301347
## class counts: 169 48
## probabilities: 0.779 0.221
## left son=1812 (43 obs) right son=1813 (174 obs)
## Primary splits:
## timbre_4_max < 115.6655 to the right, improve=1.762037, (0 missing)
## timbre_6_min < -102.3525 to the right, improve=1.140433, (0 missing)
##
## Node number 907: 25 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.4 P(node) =0.00347174
## class counts: 15 10
## probabilities: 0.600 0.400
## left son=1814 (11 obs) right son=1815 (14 obs)
## Primary splits:
## timbre_6_min < -97.937 to the left, improve=1.870130, (0 missing)
## timbre_4_max < 117.486 to the right, improve=1.333333, (0 missing)
## Surrogate splits:
## timbre_4_max < 117.1815 to the left, agree=0.64, adj=0.182, (0 split)
##
## Node number 908: 18 observations
## predicted class=N expected loss=0.05555556 P(node) =0.002499653
## class counts: 17 1
## probabilities: 0.944 0.056
##
## Node number 909: 373 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2412869 P(node) =0.05179836
## class counts: 283 90
## probabilities: 0.759 0.241
## left son=1818 (366 obs) right son=1819 (7 obs)
## Primary splits:
## timbre_4_max < 125.7355 to the right, improve=1.555094, (0 missing)
## timbre_6_min < -102.915 to the right, improve=1.013744, (0 missing)
##
## Node number 910: 19 observations
## predicted class=N expected loss=0.1578947 P(node) =0.002638522
## class counts: 16 3
## probabilities: 0.842 0.158
##
## Node number 911: 83 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3614458 P(node) =0.01152618
## class counts: 53 30
## probabilities: 0.639 0.361
## left son=1822 (72 obs) right son=1823 (11 obs)
## Primary splits:
## timbre_6_min < -101.1935 to the right, improve=1.916788, (0 missing)
## timbre_4_max < 124.599 to the left, improve=1.916788, (0 missing)
##
## Node number 918: 473 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2663848 P(node) =0.06568532
## class counts: 347 126
## probabilities: 0.734 0.266
## left son=1836 (10 obs) right son=1837 (463 obs)
## Primary splits:
## timbre_6_min < -110.4595 to the left, improve=1.449870, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=1.223526, (0 missing)
##
## Node number 919: 19 observations
## predicted class=Y expected loss=0.4736842 P(node) =0.002638522
## class counts: 9 10
## probabilities: 0.474 0.526
##
## Node number 920: 14 observations
## predicted class=N expected loss=0.07142857 P(node) =0.001944174
## class counts: 13 1
## probabilities: 0.929 0.071
##
## Node number 921: 41 observations, complexity param=0.0006289308
## predicted class=N expected loss=0.3170732 P(node) =0.005693654
## class counts: 28 13
## probabilities: 0.683 0.317
## left son=1842 (10 obs) right son=1843 (31 obs)
## Primary splits:
## timbre_4_max < 147.1105 to the left, improve=1.246420, (0 missing)
## timbre_6_min < -88.079 to the right, improve=1.092232, (0 missing)
##
## Node number 924: 7 observations
## predicted class=N expected loss=0.2857143 P(node) =0.0009720872
## class counts: 5 2
## probabilities: 0.714 0.286
##
## Node number 925: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.001805305
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 968: 17 observations
## predicted class=N expected loss=0.05882353 P(node) =0.002360783
## class counts: 16 1
## probabilities: 0.941 0.059
##
## Node number 969: 58 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2413793 P(node) =0.008054437
## class counts: 44 14
## probabilities: 0.759 0.241
## left son=1938 (30 obs) right son=1939 (28 obs)
## Primary splits:
## timbre_6_min < -124.752 to the left, improve=1.450903, (0 missing)
## timbre_4_max < 117.565 to the right, improve=1.018123, (0 missing)
## Surrogate splits:
## timbre_4_max < 123.654 to the left, agree=0.621, adj=0.214, (0 split)
##
## Node number 970: 15 observations
## predicted class=N expected loss=0.2666667 P(node) =0.002083044
## class counts: 11 4
## probabilities: 0.733 0.267
##
## Node number 971: 7 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.0009720872
## class counts: 2 5
## probabilities: 0.286 0.714
##
## Node number 974: 28 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3214286 P(node) =0.003888349
## class counts: 19 9
## probabilities: 0.679 0.321
## left son=1948 (20 obs) right son=1949 (8 obs)
## Primary splits:
## timbre_4_max < 130.9325 to the left, improve=2.0642860, (0 missing)
## timbre_6_min < -115.6625 to the right, improve=0.2610693, (0 missing)
## Surrogate splits:
## timbre_6_min < -117.0565 to the right, agree=0.857, adj=0.5, (0 split)
##
## Node number 975: 20 observations, complexity param=0.001078167
## predicted class=Y expected loss=0.4 P(node) =0.002777392
## class counts: 8 12
## probabilities: 0.400 0.600
## left son=1950 (8 obs) right son=1951 (12 obs)
## Primary splits:
## timbre_6_min < -116.3835 to the left, improve=1.350000, (0 missing)
## timbre_4_max < 113.5975 to the left, improve=0.632967, (0 missing)
## Surrogate splits:
## timbre_4_max < 112.2645 to the left, agree=0.7, adj=0.25, (0 split)
##
## Node number 980: 65 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2615385 P(node) =0.009026524
## class counts: 48 17
## probabilities: 0.738 0.262
## left son=1960 (16 obs) right son=1961 (49 obs)
## Primary splits:
## timbre_4_max < 158.734 to the right, improve=1.6816720, (0 missing)
## timbre_6_min < -118.351 to the right, improve=0.5464678, (0 missing)
## Surrogate splits:
## timbre_6_min < -114.312 to the right, agree=0.769, adj=0.063, (0 split)
##
## Node number 981: 22 observations, complexity param=0.001078167
## predicted class=N expected loss=0.4545455 P(node) =0.003055131
## class counts: 12 10
## probabilities: 0.545 0.455
## left son=1962 (11 obs) right son=1963 (11 obs)
## Primary splits:
## timbre_4_max < 180.2065 to the right, improve=1.454545, (0 missing)
## timbre_6_min < -116.6205 to the left, improve=1.051948, (0 missing)
## Surrogate splits:
## timbre_6_min < -114.0915 to the left, agree=0.682, adj=0.364, (0 split)
##
## Node number 982: 53 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3962264 P(node) =0.007360089
## class counts: 32 21
## probabilities: 0.604 0.396
## left son=1964 (32 obs) right son=1965 (21 obs)
## Primary splits:
## timbre_6_min < -130.002 to the right, improve=2.135276, (0 missing)
## timbre_4_max < 146.959 to the right, improve=1.600915, (0 missing)
##
## Node number 983: 35 observations, complexity param=0.001078167
## predicted class=Y expected loss=0.4571429 P(node) =0.004860436
## class counts: 16 19
## probabilities: 0.457 0.543
## left son=1966 (24 obs) right son=1967 (11 obs)
## Primary splits:
## timbre_6_min < -123.566 to the right, improve=2.432035, (0 missing)
## timbre_4_max < 147.26 to the left, improve=1.136134, (0 missing)
##
## Node number 1442: 885 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1299435 P(node) =0.1228996
## class counts: 770 115
## probabilities: 0.870 0.130
## left son=2884 (514 obs) right son=2885 (371 obs)
## Primary splits:
## timbre_4_max < 119.0535 to the left, improve=0.5634018, (0 missing)
## timbre_6_min < -74.981 to the right, improve=0.4190415, (0 missing)
## Surrogate splits:
## timbre_6_min < -78.0445 to the right, agree=0.592, adj=0.027, (0 split)
##
## Node number 1443: 195 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.1846154 P(node) =0.02707957
## class counts: 159 36
## probabilities: 0.815 0.185
## left son=2886 (182 obs) right son=2887 (13 obs)
## Primary splits:
## timbre_6_min < -68.378 to the right, improve=3.4879120, (0 missing)
## timbre_4_max < 110.349 to the left, improve=0.7506216, (0 missing)
##
## Node number 1448: 340 observations
## predicted class=N expected loss=0.1294118 P(node) =0.04721566
## class counts: 296 44
## probabilities: 0.871 0.129
##
## Node number 1449: 211 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1848341 P(node) =0.02930149
## class counts: 172 39
## probabilities: 0.815 0.185
## left son=2898 (123 obs) right son=2899 (88 obs)
## Primary splits:
## timbre_6_min < -80.7415 to the right, improve=1.768267, (0 missing)
## timbre_4_max < 156.9115 to the left, improve=1.105860, (0 missing)
## Surrogate splits:
## timbre_4_max < 96.6865 to the right, agree=0.592, adj=0.023, (0 split)
##
## Node number 1450: 51 observations
## predicted class=N expected loss=0.1176471 P(node) =0.00708235
## class counts: 45 6
## probabilities: 0.882 0.118
##
## Node number 1451: 12 observations
## predicted class=Y expected loss=0.3333333 P(node) =0.001666435
## class counts: 4 8
## probabilities: 0.333 0.667
##
## Node number 1452: 24 observations
## predicted class=N expected loss=0.125 P(node) =0.00333287
## class counts: 21 3
## probabilities: 0.875 0.125
##
## Node number 1453: 21 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3333333 P(node) =0.002916262
## class counts: 14 7
## probabilities: 0.667 0.333
## left son=2906 (14 obs) right son=2907 (7 obs)
## Primary splits:
## timbre_4_max < 120.2585 to the right, improve=1.1904760, (0 missing)
## timbre_6_min < -79.0535 to the right, improve=0.3888889, (0 missing)
##
## Node number 1614: 12 observations
## predicted class=N expected loss=0.1666667 P(node) =0.001666435
## class counts: 10 2
## probabilities: 0.833 0.167
##
## Node number 1615: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.001805305
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 1812: 43 observations
## predicted class=N expected loss=0.09302326 P(node) =0.005971393
## class counts: 39 4
## probabilities: 0.907 0.093
##
## Node number 1813: 174 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.2528736 P(node) =0.02416331
## class counts: 130 44
## probabilities: 0.747 0.253
## left son=3626 (81 obs) right son=3627 (93 obs)
## Primary splits:
## timbre_4_max < 113.117 to the left, improve=1.3887030, (0 missing)
## timbre_6_min < -102.898 to the right, improve=0.7858861, (0 missing)
## Surrogate splits:
## timbre_6_min < -91.4585 to the right, agree=0.563, adj=0.062, (0 split)
##
## Node number 1814: 11 observations
## predicted class=N expected loss=0.1818182 P(node) =0.001527566
## class counts: 9 2
## probabilities: 0.818 0.182
##
## Node number 1815: 14 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.001944174
## class counts: 6 8
## probabilities: 0.429 0.571
##
## Node number 1818: 366 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2349727 P(node) =0.05082627
## class counts: 280 86
## probabilities: 0.765 0.235
## left son=3636 (14 obs) right son=3637 (352 obs)
## Primary splits:
## timbre_4_max < 126.312 to the left, improve=1.607427, (0 missing)
## timbre_6_min < -102.915 to the right, improve=1.074406, (0 missing)
##
## Node number 1819: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 1822: 72 observations, complexity param=0.001078167
## predicted class=N expected loss=0.3194444 P(node) =0.009998611
## class counts: 49 23
## probabilities: 0.681 0.319
## left son=3644 (62 obs) right son=3645 (10 obs)
## Primary splits:
## timbre_4_max < 124.599 to the left, improve=3.36362, (0 missing)
## timbre_6_min < -95.6255 to the left, improve=1.46746, (0 missing)
## Surrogate splits:
## timbre_6_min < -100.4785 to the right, agree=0.875, adj=0.1, (0 split)
##
## Node number 1823: 11 observations
## predicted class=Y expected loss=0.3636364 P(node) =0.001527566
## class counts: 4 7
## probabilities: 0.364 0.636
##
## Node number 1836: 10 observations
## predicted class=N expected loss=0 P(node) =0.001388696
## class counts: 10 0
## probabilities: 1.000 0.000
##
## Node number 1837: 463 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2721382 P(node) =0.06429663
## class counts: 337 126
## probabilities: 0.728 0.272
## left son=3674 (8 obs) right son=3675 (455 obs)
## Primary splits:
## timbre_4_max < 224.1235 to the right, improve=1.205782, (0 missing)
## timbre_6_min < -92.902 to the left, improve=1.205195, (0 missing)
##
## Node number 1842: 10 observations
## predicted class=N expected loss=0.1 P(node) =0.001388696
## class counts: 9 1
## probabilities: 0.900 0.100
##
## Node number 1843: 31 observations, complexity param=0.0006289308
## predicted class=N expected loss=0.3870968 P(node) =0.004304958
## class counts: 19 12
## probabilities: 0.613 0.387
## left son=3686 (23 obs) right son=3687 (8 obs)
## Primary splits:
## timbre_6_min < -87.8225 to the right, improve=1.2205470, (0 missing)
## timbre_4_max < 173.0225 to the right, improve=0.6894754, (0 missing)
##
## Node number 1938: 30 observations
## predicted class=N expected loss=0.1333333 P(node) =0.004166088
## class counts: 26 4
## probabilities: 0.867 0.133
##
## Node number 1939: 28 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3571429 P(node) =0.003888349
## class counts: 18 10
## probabilities: 0.643 0.357
## left son=3878 (21 obs) right son=3879 (7 obs)
## Primary splits:
## timbre_4_max < 117.6805 to the right, improve=0.8571429, (0 missing)
## timbre_6_min < -123.025 to the right, improve=0.4571429, (0 missing)
## Surrogate splits:
## timbre_6_min < -120.4125 to the left, agree=0.821, adj=0.286, (0 split)
##
## Node number 1948: 20 observations
## predicted class=N expected loss=0.2 P(node) =0.002777392
## class counts: 16 4
## probabilities: 0.800 0.200
##
## Node number 1949: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 1950: 8 observations
## predicted class=N expected loss=0.375 P(node) =0.001110957
## class counts: 5 3
## probabilities: 0.625 0.375
##
## Node number 1951: 12 observations
## predicted class=Y expected loss=0.25 P(node) =0.001666435
## class counts: 3 9
## probabilities: 0.250 0.750
##
## Node number 1960: 16 observations
## predicted class=N expected loss=0.0625 P(node) =0.002221914
## class counts: 15 1
## probabilities: 0.937 0.062
##
## Node number 1961: 49 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3265306 P(node) =0.00680461
## class counts: 33 16
## probabilities: 0.673 0.327
## left son=3922 (42 obs) right son=3923 (7 obs)
## Primary splits:
## timbre_4_max < 157.723 to the left, improve=0.9795918, (0 missing)
## timbre_6_min < -119.501 to the left, improve=0.5510204, (0 missing)
##
## Node number 1962: 11 observations
## predicted class=N expected loss=0.2727273 P(node) =0.001527566
## class counts: 8 3
## probabilities: 0.727 0.273
##
## Node number 1963: 11 observations
## predicted class=Y expected loss=0.3636364 P(node) =0.001527566
## class counts: 4 7
## probabilities: 0.364 0.636
##
## Node number 1964: 32 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.28125 P(node) =0.004443827
## class counts: 23 9
## probabilities: 0.719 0.281
## left son=3928 (25 obs) right son=3929 (7 obs)
## Primary splits:
## timbre_4_max < 146.904 to the right, improve=1.5089290, (0 missing)
## timbre_6_min < -126.631 to the left, improve=0.2867063, (0 missing)
##
## Node number 1965: 21 observations, complexity param=0.001078167
## predicted class=Y expected loss=0.4285714 P(node) =0.002916262
## class counts: 9 12
## probabilities: 0.429 0.571
## left son=3930 (14 obs) right son=3931 (7 obs)
## Primary splits:
## timbre_4_max < 152.7635 to the right, improve=1.7142860, (0 missing)
## timbre_6_min < -133.339 to the left, improve=0.4285714, (0 missing)
##
## Node number 1966: 24 observations, complexity param=0.001078167
## predicted class=N expected loss=0.4166667 P(node) =0.00333287
## class counts: 14 10
## probabilities: 0.583 0.417
## left son=3932 (12 obs) right son=3933 (12 obs)
## Primary splits:
## timbre_6_min < -121.989 to the left, improve=1.3333330, (0 missing)
## timbre_4_max < 147.26 to the left, improve=0.3333333, (0 missing)
## Surrogate splits:
## timbre_4_max < 156.103 to the left, agree=0.625, adj=0.25, (0 split)
##
## Node number 1967: 11 observations
## predicted class=Y expected loss=0.1818182 P(node) =0.001527566
## class counts: 2 9
## probabilities: 0.182 0.818
##
## Node number 2884: 514 observations
## predicted class=N expected loss=0.114786 P(node) =0.07137898
## class counts: 455 59
## probabilities: 0.885 0.115
##
## Node number 2885: 371 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1509434 P(node) =0.05152062
## class counts: 315 56
## probabilities: 0.849 0.151
## left son=5770 (360 obs) right son=5771 (11 obs)
## Primary splits:
## timbre_4_max < 119.9485 to the right, improve=2.089794, (0 missing)
## timbre_6_min < -78.103 to the left, improve=1.761006, (0 missing)
##
## Node number 2886: 182 observations
## predicted class=N expected loss=0.1593407 P(node) =0.02527427
## class counts: 153 29
## probabilities: 0.841 0.159
##
## Node number 2887: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.001805305
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 2898: 123 observations
## predicted class=N expected loss=0.1300813 P(node) =0.01708096
## class counts: 107 16
## probabilities: 0.870 0.130
##
## Node number 2899: 88 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2613636 P(node) =0.01222052
## class counts: 65 23
## probabilities: 0.739 0.261
## left son=5798 (81 obs) right son=5799 (7 obs)
## Primary splits:
## timbre_6_min < -80.8265 to the left, improve=1.4622820, (0 missing)
## timbre_4_max < 125.668 to the right, improve=0.4402042, (0 missing)
##
## Node number 2906: 14 observations
## predicted class=N expected loss=0.2142857 P(node) =0.001944174
## class counts: 11 3
## probabilities: 0.786 0.214
##
## Node number 2907: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 3626: 81 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.1851852 P(node) =0.01124844
## class counts: 66 15
## probabilities: 0.815 0.185
## left son=7252 (11 obs) right son=7253 (70 obs)
## Primary splits:
## timbre_6_min < -89.441 to the right, improve=0.8730159, (0 missing)
## timbre_4_max < 110.7735 to the right, improve=0.6463675, (0 missing)
##
## Node number 3627: 93 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.311828 P(node) =0.01291487
## class counts: 64 29
## probabilities: 0.688 0.312
## left son=7254 (66 obs) right son=7255 (27 obs)
## Primary splits:
## timbre_6_min < -102.816 to the right, improve=1.338221, (0 missing)
## timbre_4_max < 113.891 to the right, improve=1.292766, (0 missing)
##
## Node number 3636: 14 observations
## predicted class=N expected loss=0 P(node) =0.001944174
## class counts: 14 0
## probabilities: 1.000 0.000
##
## Node number 3637: 352 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2443182 P(node) =0.0488821
## class counts: 266 86
## probabilities: 0.756 0.244
## left son=7274 (254 obs) right son=7275 (98 obs)
## Primary splits:
## timbre_6_min < -102.915 to the right, improve=1.0375330, (0 missing)
## timbre_4_max < 127.3095 to the right, improve=0.7744509, (0 missing)
##
## Node number 3644: 62 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.2580645 P(node) =0.008609915
## class counts: 46 16
## probabilities: 0.742 0.258
## left son=7288 (21 obs) right son=7289 (41 obs)
## Primary splits:
## timbre_4_max < 123.629 to the right, improve=0.8429808, (0 missing)
## timbre_6_min < -98.7895 to the right, improve=0.4803970, (0 missing)
##
## Node number 3645: 10 observations
## predicted class=Y expected loss=0.3 P(node) =0.001388696
## class counts: 3 7
## probabilities: 0.300 0.700
##
## Node number 3674: 8 observations
## predicted class=N expected loss=0 P(node) =0.001110957
## class counts: 8 0
## probabilities: 1.000 0.000
##
## Node number 3675: 455 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2769231 P(node) =0.06318567
## class counts: 329 126
## probabilities: 0.723 0.277
## left son=7350 (401 obs) right son=7351 (54 obs)
## Primary splits:
## timbre_6_min < -92.902 to the left, improve=1.5362490, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.8936559, (0 missing)
##
## Node number 3686: 23 observations, complexity param=0.0006289308
## predicted class=N expected loss=0.3043478 P(node) =0.003194001
## class counts: 16 7
## probabilities: 0.696 0.304
## left son=7372 (14 obs) right son=7373 (9 obs)
## Primary splits:
## timbre_4_max < 151.452 to the right, improve=1.866115, (0 missing)
## timbre_6_min < -86.75 to the left, improve=0.310559, (0 missing)
## Surrogate splits:
## timbre_6_min < -87.318 to the left, agree=0.652, adj=0.111, (0 split)
##
## Node number 3687: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 3878: 21 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2857143 P(node) =0.002916262
## class counts: 15 6
## probabilities: 0.714 0.286
## left son=7756 (12 obs) right son=7757 (9 obs)
## Primary splits:
## timbre_6_min < -122.317 to the right, improve=2.2936510, (0 missing)
## timbre_4_max < 124.098 to the left, improve=0.9603175, (0 missing)
## Surrogate splits:
## timbre_4_max < 131.217 to the right, agree=0.667, adj=0.222, (0 split)
##
## Node number 3879: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 3922: 42 observations
## predicted class=N expected loss=0.2857143 P(node) =0.005832523
## class counts: 30 12
## probabilities: 0.714 0.286
##
## Node number 3923: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 3928: 25 observations
## predicted class=N expected loss=0.2 P(node) =0.00347174
## class counts: 20 5
## probabilities: 0.800 0.200
##
## Node number 3929: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 3930: 14 observations
## predicted class=N expected loss=0.4285714 P(node) =0.001944174
## class counts: 8 6
## probabilities: 0.571 0.429
##
## Node number 3931: 7 observations
## predicted class=Y expected loss=0.1428571 P(node) =0.0009720872
## class counts: 1 6
## probabilities: 0.143 0.857
##
## Node number 3932: 12 observations
## predicted class=N expected loss=0.25 P(node) =0.001666435
## class counts: 9 3
## probabilities: 0.750 0.250
##
## Node number 3933: 12 observations
## predicted class=Y expected loss=0.4166667 P(node) =0.001666435
## class counts: 5 7
## probabilities: 0.417 0.583
##
## Node number 5770: 360 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1416667 P(node) =0.04999306
## class counts: 309 51
## probabilities: 0.858 0.142
## left son=11540 (35 obs) right son=11541 (325 obs)
## Primary splits:
## timbre_6_min < -78.103 to the left, improve=1.5561540, (0 missing)
## timbre_4_max < 124.9375 to the right, improve=0.4514314, (0 missing)
##
## Node number 5771: 11 observations
## predicted class=N expected loss=0.4545455 P(node) =0.001527566
## class counts: 6 5
## probabilities: 0.545 0.455
##
## Node number 5798: 81 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2345679 P(node) =0.01124844
## class counts: 62 19
## probabilities: 0.765 0.235
## left son=11596 (58 obs) right son=11597 (23 obs)
## Primary splits:
## timbre_6_min < -81.383 to the right, improve=0.8240509, (0 missing)
## timbre_4_max < 103.923 to the right, improve=0.6244479, (0 missing)
## Surrogate splits:
## timbre_4_max < 98.883 to the right, agree=0.728, adj=0.043, (0 split)
##
## Node number 5799: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 7252: 11 observations
## predicted class=N expected loss=0 P(node) =0.001527566
## class counts: 11 0
## probabilities: 1.000 0.000
##
## Node number 7253: 70 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.2142857 P(node) =0.009720872
## class counts: 55 15
## probabilities: 0.786 0.214
## left son=14506 (46 obs) right son=14507 (24 obs)
## Primary splits:
## timbre_6_min < -93.8255 to the left, improve=1.0351970, (0 missing)
## timbre_4_max < 110.7735 to the right, improve=0.7142857, (0 missing)
##
## Node number 7254: 66 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.2575758 P(node) =0.009165394
## class counts: 49 17
## probabilities: 0.742 0.258
## left son=14508 (26 obs) right son=14509 (40 obs)
## Primary splits:
## timbre_6_min < -95.7075 to the left, improve=1.7347320, (0 missing)
## timbre_4_max < 114.5965 to the right, improve=0.9231935, (0 missing)
##
## Node number 7255: 27 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.4444444 P(node) =0.003749479
## class counts: 15 12
## probabilities: 0.556 0.444
## left son=14510 (17 obs) right son=14511 (10 obs)
## Primary splits:
## timbre_6_min < -105.5865 to the left, improve=2.0745100, (0 missing)
## timbre_4_max < 114.859 to the right, improve=0.5333333, (0 missing)
## Surrogate splits:
## timbre_4_max < 113.6145 to the right, agree=0.741, adj=0.3, (0 split)
##
## Node number 7274: 254 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2204724 P(node) =0.03527288
## class counts: 198 56
## probabilities: 0.780 0.220
## left son=14548 (195 obs) right son=14549 (59 obs)
## Primary splits:
## timbre_4_max < 134.869 to the left, improve=1.585400, (0 missing)
## timbre_6_min < -101.875 to the left, improve=1.117786, (0 missing)
##
## Node number 7275: 98 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3061224 P(node) =0.01360922
## class counts: 68 30
## probabilities: 0.694 0.306
## left son=14550 (87 obs) right son=14551 (11 obs)
## Primary splits:
## timbre_4_max < 127.5305 to the right, improve=1.419487, (0 missing)
## timbre_6_min < -108.277 to the left, improve=1.090139, (0 missing)
##
## Node number 7288: 21 observations
## predicted class=N expected loss=0.1428571 P(node) =0.002916262
## class counts: 18 3
## probabilities: 0.857 0.143
##
## Node number 7289: 41 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.3170732 P(node) =0.005693654
## class counts: 28 13
## probabilities: 0.683 0.317
## left son=14578 (18 obs) right son=14579 (23 obs)
## Primary splits:
## timbre_6_min < -96.115 to the left, improve=0.5773536, (0 missing)
## timbre_4_max < 122.467 to the left, improve=0.2360976, (0 missing)
## Surrogate splits:
## timbre_4_max < 122.499 to the left, agree=0.683, adj=0.278, (0 split)
##
## Node number 7350: 401 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2618454 P(node) =0.05568671
## class counts: 296 105
## probabilities: 0.738 0.262
## left son=14700 (22 obs) right son=14701 (379 obs)
## Primary splits:
## timbre_6_min < -93.8975 to the right, improve=1.3602740, (0 missing)
## timbre_4_max < 187.1575 to the left, improve=0.6537155, (0 missing)
##
## Node number 7351: 54 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.3888889 P(node) =0.007498958
## class counts: 33 21
## probabilities: 0.611 0.389
## left son=14702 (44 obs) right son=14703 (10 obs)
## Primary splits:
## timbre_4_max < 141.9325 to the right, improve=2.3757580, (0 missing)
## timbre_6_min < -92.3515 to the right, improve=0.7843137, (0 missing)
##
## Node number 7372: 14 observations
## predicted class=N expected loss=0.1428571 P(node) =0.001944174
## class counts: 12 2
## probabilities: 0.857 0.143
##
## Node number 7373: 9 observations
## predicted class=Y expected loss=0.4444444 P(node) =0.001249826
## class counts: 4 5
## probabilities: 0.444 0.556
##
## Node number 7756: 12 observations
## predicted class=N expected loss=0.08333333 P(node) =0.001666435
## class counts: 11 1
## probabilities: 0.917 0.083
##
## Node number 7757: 9 observations
## predicted class=Y expected loss=0.4444444 P(node) =0.001249826
## class counts: 4 5
## probabilities: 0.444 0.556
##
## Node number 11540: 35 observations
## predicted class=N expected loss=0 P(node) =0.004860436
## class counts: 35 0
## probabilities: 1.000 0.000
##
## Node number 11541: 325 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1569231 P(node) =0.04513262
## class counts: 274 51
## probabilities: 0.843 0.157
## left son=23082 (314 obs) right son=23083 (11 obs)
## Primary splits:
## timbre_6_min < -77.8695 to the right, improve=2.0170080, (0 missing)
## timbre_4_max < 120.4265 to the left, improve=0.4558715, (0 missing)
##
## Node number 11596: 58 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.1896552 P(node) =0.008054437
## class counts: 47 11
## probabilities: 0.810 0.190
## left son=23192 (24 obs) right son=23193 (34 obs)
## Primary splits:
## timbre_6_min < -81.14 to the left, improve=1.7932720, (0 missing)
## timbre_4_max < 155.0395 to the left, improve=0.9088187, (0 missing)
## Surrogate splits:
## timbre_4_max < 110.414 to the left, agree=0.603, adj=0.042, (0 split)
##
## Node number 11597: 23 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.3478261 P(node) =0.003194001
## class counts: 15 8
## probabilities: 0.652 0.348
## left son=23194 (16 obs) right son=23195 (7 obs)
## Primary splits:
## timbre_6_min < -81.452 to the left, improve=1.0062110, (0 missing)
## timbre_4_max < 125.668 to the right, improve=0.8454969, (0 missing)
##
## Node number 14506: 46 observations
## predicted class=N expected loss=0.1521739 P(node) =0.006388002
## class counts: 39 7
## probabilities: 0.848 0.152
##
## Node number 14507: 24 observations, complexity param=0.0003144654
## predicted class=N expected loss=0.3333333 P(node) =0.00333287
## class counts: 16 8
## probabilities: 0.667 0.333
## left son=29014 (17 obs) right son=29015 (7 obs)
## Primary splits:
## timbre_4_max < 111.553 to the right, improve=1.1204480, (0 missing)
## timbre_6_min < -90.4815 to the left, improve=0.1792717, (0 missing)
## Surrogate splits:
## timbre_6_min < -93.2075 to the right, agree=0.792, adj=0.286, (0 split)
##
## Node number 14508: 26 observations
## predicted class=N expected loss=0.1153846 P(node) =0.00361061
## class counts: 23 3
## probabilities: 0.885 0.115
##
## Node number 14509: 40 observations, complexity param=0.0008086253
## predicted class=N expected loss=0.35 P(node) =0.005554784
## class counts: 26 14
## probabilities: 0.650 0.350
## left son=29018 (30 obs) right son=29019 (10 obs)
## Primary splits:
## timbre_4_max < 113.5145 to the right, improve=1.6666670, (0 missing)
## timbre_6_min < -92.434 to the right, improve=0.6533333, (0 missing)
##
## Node number 14510: 17 observations
## predicted class=N expected loss=0.2941176 P(node) =0.002360783
## class counts: 12 5
## probabilities: 0.706 0.294
##
## Node number 14511: 10 observations
## predicted class=Y expected loss=0.3 P(node) =0.001388696
## class counts: 3 7
## probabilities: 0.300 0.700
##
## Node number 14548: 195 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.1897436 P(node) =0.02707957
## class counts: 158 37
## probabilities: 0.810 0.190
## left son=29096 (47 obs) right son=29097 (148 obs)
## Primary splits:
## timbre_4_max < 133.052 to the right, improve=1.3560420, (0 missing)
## timbre_6_min < -100.7345 to the left, improve=0.7141533, (0 missing)
## Surrogate splits:
## timbre_6_min < -86.286 to the right, agree=0.764, adj=0.021, (0 split)
##
## Node number 14549: 59 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3220339 P(node) =0.008193306
## class counts: 40 19
## probabilities: 0.678 0.322
## left son=29098 (28 obs) right son=29099 (31 obs)
## Primary splits:
## timbre_6_min < -93.3205 to the right, improve=1.2373660, (0 missing)
## timbre_4_max < 137.2545 to the right, improve=0.7185942, (0 missing)
## Surrogate splits:
## timbre_4_max < 137.2775 to the right, agree=0.61, adj=0.179, (0 split)
##
## Node number 14550: 87 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2758621 P(node) =0.01208166
## class counts: 63 24
## probabilities: 0.724 0.276
## left son=29100 (24 obs) right son=29101 (63 obs)
## Primary splits:
## timbre_6_min < -108.277 to the left, improve=0.7903667, (0 missing)
## timbre_4_max < 130.4925 to the left, improve=0.6499250, (0 missing)
##
## Node number 14551: 11 observations
## predicted class=Y expected loss=0.4545455 P(node) =0.001527566
## class counts: 5 6
## probabilities: 0.455 0.545
##
## Node number 14578: 18 observations
## predicted class=N expected loss=0.2222222 P(node) =0.002499653
## class counts: 14 4
## probabilities: 0.778 0.222
##
## Node number 14579: 23 observations, complexity param=0.0009433962
## predicted class=N expected loss=0.3913043 P(node) =0.003194001
## class counts: 14 9
## probabilities: 0.609 0.391
## left son=29158 (14 obs) right son=29159 (9 obs)
## Primary splits:
## timbre_6_min < -93.072 to the right, improve=2.242236, (0 missing)
## timbre_4_max < 122.9475 to the right, improve=1.294983, (0 missing)
## Surrogate splits:
## timbre_4_max < 122.9475 to the right, agree=0.652, adj=0.111, (0 split)
##
## Node number 14700: 22 observations
## predicted class=N expected loss=0.09090909 P(node) =0.003055131
## class counts: 20 2
## probabilities: 0.909 0.091
##
## Node number 14701: 379 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2717678 P(node) =0.05263158
## class counts: 276 103
## probabilities: 0.728 0.272
## left son=29402 (357 obs) right son=29403 (22 obs)
## Primary splits:
## timbre_4_max < 187.1575 to the left, improve=0.8808681, (0 missing)
## timbre_6_min < -100.37 to the left, improve=0.7616642, (0 missing)
##
## Node number 14702: 44 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.3181818 P(node) =0.006110262
## class counts: 30 14
## probabilities: 0.682 0.318
## left son=29404 (15 obs) right son=29405 (29 obs)
## Primary splits:
## timbre_4_max < 150.4675 to the left, improve=2.879415, (0 missing)
## timbre_6_min < -92.3515 to the right, improve=1.286988, (0 missing)
##
## Node number 14703: 10 observations
## predicted class=Y expected loss=0.3 P(node) =0.001388696
## class counts: 3 7
## probabilities: 0.300 0.700
##
## Node number 23082: 314 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1464968 P(node) =0.04360505
## class counts: 268 46
## probabilities: 0.854 0.146
## left son=46164 (76 obs) right son=46165 (238 obs)
## Primary splits:
## timbre_6_min < -76.0415 to the left, improve=0.5932793, (0 missing)
## timbre_4_max < 121.5315 to the right, improve=0.4026349, (0 missing)
## Surrogate splits:
## timbre_4_max < 185.1995 to the right, agree=0.764, adj=0.026, (0 split)
##
## Node number 23083: 11 observations
## predicted class=N expected loss=0.4545455 P(node) =0.001527566
## class counts: 6 5
## probabilities: 0.545 0.455
##
## Node number 23192: 24 observations
## predicted class=N expected loss=0.04166667 P(node) =0.00333287
## class counts: 23 1
## probabilities: 0.958 0.042
##
## Node number 23193: 34 observations, complexity param=0.0005390836
## predicted class=N expected loss=0.2941176 P(node) =0.004721566
## class counts: 24 10
## probabilities: 0.706 0.294
## left son=46386 (27 obs) right son=46387 (7 obs)
## Primary splits:
## timbre_6_min < -81.1045 to the right, improve=3.1123560, (0 missing)
## timbre_4_max < 142.4145 to the left, improve=0.3187053, (0 missing)
## Surrogate splits:
## timbre_4_max < 147.683 to the left, agree=0.912, adj=0.571, (0 split)
##
## Node number 23194: 16 observations
## predicted class=N expected loss=0.25 P(node) =0.002221914
## class counts: 12 4
## probabilities: 0.750 0.250
##
## Node number 23195: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 29014: 17 observations
## predicted class=N expected loss=0.2352941 P(node) =0.002360783
## class counts: 13 4
## probabilities: 0.765 0.235
##
## Node number 29015: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 29018: 30 observations
## predicted class=N expected loss=0.2666667 P(node) =0.004166088
## class counts: 22 8
## probabilities: 0.733 0.267
##
## Node number 29019: 10 observations
## predicted class=Y expected loss=0.4 P(node) =0.001388696
## class counts: 4 6
## probabilities: 0.400 0.600
##
## Node number 29096: 47 observations
## predicted class=N expected loss=0.08510638 P(node) =0.006526871
## class counts: 43 4
## probabilities: 0.915 0.085
##
## Node number 29097: 148 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.222973 P(node) =0.0205527
## class counts: 115 33
## probabilities: 0.777 0.223
## left son=58194 (73 obs) right son=58195 (75 obs)
## Primary splits:
## timbre_4_max < 129.356 to the left, improve=1.505519, (0 missing)
## timbre_6_min < -99.202 to the left, improve=1.035019, (0 missing)
## Surrogate splits:
## timbre_6_min < -93.21 to the right, agree=0.595, adj=0.178, (0 split)
##
## Node number 29098: 28 observations
## predicted class=N expected loss=0.2142857 P(node) =0.003888349
## class counts: 22 6
## probabilities: 0.786 0.214
##
## Node number 29099: 31 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.4193548 P(node) =0.004304958
## class counts: 18 13
## probabilities: 0.581 0.419
## left son=58198 (19 obs) right son=58199 (12 obs)
## Primary splits:
## timbre_4_max < 135.773 to the right, improve=1.0529150, (0 missing)
## timbre_6_min < -100.434 to the right, improve=0.4182028, (0 missing)
## Surrogate splits:
## timbre_6_min < -93.773 to the left, agree=0.677, adj=0.167, (0 split)
##
## Node number 29100: 24 observations
## predicted class=N expected loss=0.1666667 P(node) =0.00333287
## class counts: 20 4
## probabilities: 0.833 0.167
##
## Node number 29101: 63 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3174603 P(node) =0.008748785
## class counts: 43 20
## probabilities: 0.683 0.317
## left son=58202 (18 obs) right son=58203 (45 obs)
## Primary splits:
## timbre_4_max < 130.4925 to the left, improve=1.1460320, (0 missing)
## timbre_6_min < -107.592 to the right, improve=0.3386243, (0 missing)
## Surrogate splits:
## timbre_6_min < -107.671 to the left, agree=0.746, adj=0.111, (0 split)
##
## Node number 29158: 14 observations
## predicted class=N expected loss=0.2142857 P(node) =0.001944174
## class counts: 11 3
## probabilities: 0.786 0.214
##
## Node number 29159: 9 observations
## predicted class=Y expected loss=0.3333333 P(node) =0.001249826
## class counts: 3 6
## probabilities: 0.333 0.667
##
## Node number 29402: 357 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.2633053 P(node) =0.04957645
## class counts: 263 94
## probabilities: 0.737 0.263
## left son=58804 (288 obs) right son=58805 (69 obs)
## Primary splits:
## timbre_6_min < -96.643 to the left, improve=0.8388772, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.6154952, (0 missing)
##
## Node number 29403: 22 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.4090909 P(node) =0.003055131
## class counts: 13 9
## probabilities: 0.591 0.409
## left son=58806 (15 obs) right son=58807 (7 obs)
## Primary splits:
## timbre_4_max < 195.063 to the right, improve=1.912554, (0 missing)
## timbre_6_min < -99.7295 to the right, improve=1.063714, (0 missing)
##
## Node number 29404: 15 observations
## predicted class=N expected loss=0.06666667 P(node) =0.002083044
## class counts: 14 1
## probabilities: 0.933 0.067
##
## Node number 29405: 29 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.4482759 P(node) =0.004027218
## class counts: 16 13
## probabilities: 0.552 0.448
## left son=58810 (8 obs) right son=58811 (21 obs)
## Primary splits:
## timbre_4_max < 174.053 to the right, improve=2.309113, (0 missing)
## timbre_6_min < -92.3515 to the right, improve=1.952671, (0 missing)
##
## Node number 46164: 76 observations
## predicted class=N expected loss=0.09210526 P(node) =0.01055409
## class counts: 69 7
## probabilities: 0.908 0.092
##
## Node number 46165: 238 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1638655 P(node) =0.03305097
## class counts: 199 39
## probabilities: 0.836 0.164
## left son=92330 (204 obs) right son=92331 (34 obs)
## Primary splits:
## timbre_6_min < -75.1125 to the right, improve=1.3459380, (0 missing)
## timbre_4_max < 120.4265 to the left, improve=0.5023302, (0 missing)
##
## Node number 46386: 27 observations
## predicted class=N expected loss=0.1851852 P(node) =0.003749479
## class counts: 22 5
## probabilities: 0.815 0.185
##
## Node number 46387: 7 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.0009720872
## class counts: 2 5
## probabilities: 0.286 0.714
##
## Node number 58194: 73 observations
## predicted class=N expected loss=0.1506849 P(node) =0.01013748
## class counts: 62 11
## probabilities: 0.849 0.151
##
## Node number 58195: 75 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2933333 P(node) =0.01041522
## class counts: 53 22
## probabilities: 0.707 0.293
## left son=116390 (67 obs) right son=116391 (8 obs)
## Primary splits:
## timbre_4_max < 129.608 to the right, improve=3.735124, (0 missing)
## timbre_6_min < -95.8915 to the left, improve=1.548592, (0 missing)
##
## Node number 58198: 19 observations
## predicted class=N expected loss=0.3157895 P(node) =0.002638522
## class counts: 13 6
## probabilities: 0.684 0.316
##
## Node number 58199: 12 observations
## predicted class=Y expected loss=0.4166667 P(node) =0.001666435
## class counts: 5 7
## probabilities: 0.417 0.583
##
## Node number 58202: 18 observations
## predicted class=N expected loss=0.1666667 P(node) =0.002499653
## class counts: 15 3
## probabilities: 0.833 0.167
##
## Node number 58203: 45 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3777778 P(node) =0.006249132
## class counts: 28 17
## probabilities: 0.622 0.378
## left son=116406 (29 obs) right son=116407 (16 obs)
## Primary splits:
## timbre_4_max < 132.712 to the right, improve=3.034866, (0 missing)
## timbre_6_min < -104.2985 to the left, improve=0.384127, (0 missing)
## Surrogate splits:
## timbre_6_min < -107.516 to the right, agree=0.667, adj=0.062, (0 split)
##
## Node number 58804: 288 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2465278 P(node) =0.03999445
## class counts: 217 71
## probabilities: 0.753 0.247
## left son=117608 (9 obs) right son=117609 (279 obs)
## Primary splits:
## timbre_6_min < -97.0515 to the right, improve=1.1292560, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.7162698, (0 missing)
##
## Node number 58805: 69 observations, complexity param=0.0006738544
## predicted class=N expected loss=0.3333333 P(node) =0.009582002
## class counts: 46 23
## probabilities: 0.667 0.333
## left son=117610 (55 obs) right son=117611 (14 obs)
## Primary splits:
## timbre_6_min < -96.346 to the right, improve=5.097835, (0 missing)
## timbre_4_max < 140.693 to the left, improve=1.731183, (0 missing)
##
## Node number 58806: 15 observations
## predicted class=N expected loss=0.2666667 P(node) =0.002083044
## class counts: 11 4
## probabilities: 0.733 0.267
##
## Node number 58807: 7 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.0009720872
## class counts: 2 5
## probabilities: 0.286 0.714
##
## Node number 58810: 8 observations
## predicted class=N expected loss=0.125 P(node) =0.001110957
## class counts: 7 1
## probabilities: 0.875 0.125
##
## Node number 58811: 21 observations, complexity param=0.0006738544
## predicted class=Y expected loss=0.4285714 P(node) =0.002916262
## class counts: 9 12
## probabilities: 0.429 0.571
## left son=117622 (12 obs) right son=117623 (9 obs)
## Primary splits:
## timbre_6_min < -92.31 to the right, improve=1.3412700, (0 missing)
## timbre_4_max < 160.1845 to the right, improve=0.4285714, (0 missing)
## Surrogate splits:
## timbre_4_max < 163.2825 to the left, agree=0.667, adj=0.222, (0 split)
##
## Node number 92330: 204 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1421569 P(node) =0.0283294
## class counts: 175 29
## probabilities: 0.858 0.142
## left son=184660 (12 obs) right son=184661 (192 obs)
## Primary splits:
## timbre_4_max < 120.8475 to the left, improve=0.5153186, (0 missing)
## timbre_6_min < -70.573 to the right, improve=0.2428566, (0 missing)
##
## Node number 92331: 34 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.2941176 P(node) =0.004721566
## class counts: 24 10
## probabilities: 0.706 0.294
## left son=184662 (13 obs) right son=184663 (21 obs)
## Primary splits:
## timbre_6_min < -75.681 to the left, improve=0.8282698, (0 missing)
## timbre_4_max < 151.6005 to the right, improve=0.4033613, (0 missing)
## Surrogate splits:
## timbre_4_max < 147.3055 to the right, agree=0.706, adj=0.231, (0 split)
##
## Node number 116390: 67 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.238806 P(node) =0.009304263
## class counts: 51 16
## probabilities: 0.761 0.239
## left son=232780 (22 obs) right son=232781 (45 obs)
## Primary splits:
## timbre_6_min < -95.8915 to the left, improve=2.449118, (0 missing)
## timbre_4_max < 130.831 to the left, improve=1.306738, (0 missing)
##
## Node number 116391: 8 observations
## predicted class=Y expected loss=0.25 P(node) =0.001110957
## class counts: 2 6
## probabilities: 0.250 0.750
##
## Node number 116406: 29 observations
## predicted class=N expected loss=0.2413793 P(node) =0.004027218
## class counts: 22 7
## probabilities: 0.759 0.241
##
## Node number 116407: 16 observations
## predicted class=Y expected loss=0.375 P(node) =0.002221914
## class counts: 6 10
## probabilities: 0.375 0.625
##
## Node number 117608: 9 observations
## predicted class=N expected loss=0 P(node) =0.001249826
## class counts: 9 0
## probabilities: 1.000 0.000
##
## Node number 117609: 279 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2544803 P(node) =0.03874462
## class counts: 208 71
## probabilities: 0.746 0.254
## left son=235218 (272 obs) right son=235219 (7 obs)
## Primary splits:
## timbre_6_min < -97.331 to the left, improve=1.4425810, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.6455453, (0 missing)
##
## Node number 117610: 55 observations
## predicted class=N expected loss=0.2363636 P(node) =0.007637828
## class counts: 42 13
## probabilities: 0.764 0.236
##
## Node number 117611: 14 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.001944174
## class counts: 4 10
## probabilities: 0.286 0.714
##
## Node number 117622: 12 observations
## predicted class=N expected loss=0.4166667 P(node) =0.001666435
## class counts: 7 5
## probabilities: 0.583 0.417
##
## Node number 117623: 9 observations
## predicted class=Y expected loss=0.2222222 P(node) =0.001249826
## class counts: 2 7
## probabilities: 0.222 0.778
##
## Node number 184660: 12 observations
## predicted class=N expected loss=0 P(node) =0.001666435
## class counts: 12 0
## probabilities: 1.000 0.000
##
## Node number 184661: 192 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1510417 P(node) =0.02666296
## class counts: 163 29
## probabilities: 0.849 0.151
## left son=369322 (185 obs) right son=369323 (7 obs)
## Primary splits:
## timbre_4_max < 121.6295 to the right, improve=1.1191200, (0 missing)
## timbre_6_min < -70.573 to the right, improve=0.3047094, (0 missing)
##
## Node number 184662: 13 observations
## predicted class=N expected loss=0.1538462 P(node) =0.001805305
## class counts: 11 2
## probabilities: 0.846 0.154
##
## Node number 184663: 21 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.3809524 P(node) =0.002916262
## class counts: 13 8
## probabilities: 0.619 0.381
## left son=369326 (14 obs) right son=369327 (7 obs)
## Primary splits:
## timbre_6_min < -75.4695 to the right, improve=0.7619048, (0 missing)
## timbre_4_max < 131.742 to the right, improve=0.7619048, (0 missing)
## Surrogate splits:
## timbre_4_max < 137.361 to the left, agree=0.714, adj=0.143, (0 split)
##
## Node number 232780: 22 observations
## predicted class=N expected loss=0.04545455 P(node) =0.003055131
## class counts: 21 1
## probabilities: 0.955 0.045
##
## Node number 232781: 45 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3333333 P(node) =0.006249132
## class counts: 30 15
## probabilities: 0.667 0.333
## left son=465562 (38 obs) right son=465563 (7 obs)
## Primary splits:
## timbre_4_max < 132.4285 to the left, improve=2.406015, (0 missing)
## timbre_6_min < -94.821 to the right, improve=1.655405, (0 missing)
##
## Node number 235218: 272 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2463235 P(node) =0.03777253
## class counts: 205 67
## probabilities: 0.754 0.246
## left son=470436 (9 obs) right son=470437 (263 obs)
## Primary splits:
## timbre_6_min < -97.639 to the right, improve=1.1295290, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.6675752, (0 missing)
##
## Node number 235219: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 369322: 185 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1405405 P(node) =0.02569088
## class counts: 159 26
## probabilities: 0.859 0.141
## left son=738644 (16 obs) right son=738645 (169 obs)
## Primary splits:
## timbre_4_max < 123.7945 to the left, improve=0.6918919, (0 missing)
## timbre_6_min < -74.8875 to the left, improve=0.4176062, (0 missing)
##
## Node number 369323: 7 observations
## predicted class=N expected loss=0.4285714 P(node) =0.0009720872
## class counts: 4 3
## probabilities: 0.571 0.429
##
## Node number 369326: 14 observations
## predicted class=N expected loss=0.2857143 P(node) =0.001944174
## class counts: 10 4
## probabilities: 0.714 0.286
##
## Node number 369327: 7 observations
## predicted class=Y expected loss=0.4285714 P(node) =0.0009720872
## class counts: 3 4
## probabilities: 0.429 0.571
##
## Node number 465562: 38 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2631579 P(node) =0.005277045
## class counts: 28 10
## probabilities: 0.737 0.263
## left son=931124 (12 obs) right son=931125 (26 obs)
## Primary splits:
## timbre_4_max < 130.831 to the left, improve=1.1342780, (0 missing)
## timbre_6_min < -89.7845 to the right, improve=0.4722267, (0 missing)
##
## Node number 465563: 7 observations
## predicted class=Y expected loss=0.2857143 P(node) =0.0009720872
## class counts: 2 5
## probabilities: 0.286 0.714
##
## Node number 470436: 9 observations
## predicted class=N expected loss=0 P(node) =0.001249826
## class counts: 9 0
## probabilities: 1.000 0.000
##
## Node number 470437: 263 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2547529 P(node) =0.03652271
## class counts: 196 67
## probabilities: 0.745 0.255
## left son=940874 (243 obs) right son=940875 (20 obs)
## Primary splits:
## timbre_6_min < -109.2095 to the right, improve=0.9133236, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.6886413, (0 missing)
##
## Node number 738644: 16 observations
## predicted class=N expected loss=0 P(node) =0.002221914
## class counts: 16 0
## probabilities: 1.000 0.000
##
## Node number 738645: 169 observations, complexity param=0.0002358491
## predicted class=N expected loss=0.1538462 P(node) =0.02346896
## class counts: 143 26
## probabilities: 0.846 0.154
## left son=1477290 (161 obs) right son=1477291 (8 obs)
## Primary splits:
## timbre_4_max < 124.9935 to the right, improve=3.728261, (0 missing)
## timbre_6_min < -74.8875 to the left, improve=0.450000, (0 missing)
##
## Node number 931124: 12 observations
## predicted class=N expected loss=0.08333333 P(node) =0.001666435
## class counts: 11 1
## probabilities: 0.917 0.083
##
## Node number 931125: 26 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3461538 P(node) =0.00361061
## class counts: 17 9
## probabilities: 0.654 0.346
## left son=1862250 (13 obs) right son=1862251 (13 obs)
## Primary splits:
## timbre_4_max < 131.897 to the right, improve=1.9230770, (0 missing)
## timbre_6_min < -89.763 to the right, improve=0.2136752, (0 missing)
## Surrogate splits:
## timbre_6_min < -90.37 to the right, agree=0.654, adj=0.308, (0 split)
##
## Node number 940874: 243 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2427984 P(node) =0.03374531
## class counts: 184 59
## probabilities: 0.757 0.243
## left son=1881748 (38 obs) right son=1881749 (205 obs)
## Primary splits:
## timbre_6_min < -107.1325 to the left, improve=1.1143640, (0 missing)
## timbre_4_max < 144.1545 to the right, improve=0.8998324, (0 missing)
##
## Node number 940875: 20 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.4 P(node) =0.002777392
## class counts: 12 8
## probabilities: 0.600 0.400
## left son=1881750 (9 obs) right son=1881751 (11 obs)
## Primary splits:
## timbre_4_max < 149.5945 to the left, improve=1.0343430, (0 missing)
## timbre_6_min < -110.0255 to the left, improve=0.1454545, (0 missing)
## Surrogate splits:
## timbre_6_min < -110.1945 to the right, agree=0.65, adj=0.222, (0 split)
##
## Node number 1477290: 161 observations
## predicted class=N expected loss=0.1304348 P(node) =0.02235801
## class counts: 140 21
## probabilities: 0.870 0.130
##
## Node number 1477291: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 1862250: 13 observations
## predicted class=N expected loss=0.1538462 P(node) =0.001805305
## class counts: 11 2
## probabilities: 0.846 0.154
##
## Node number 1862251: 13 observations
## predicted class=Y expected loss=0.4615385 P(node) =0.001805305
## class counts: 6 7
## probabilities: 0.462 0.538
##
## Node number 1881748: 38 observations
## predicted class=N expected loss=0.1315789 P(node) =0.005277045
## class counts: 33 5
## probabilities: 0.868 0.132
##
## Node number 1881749: 205 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2634146 P(node) =0.02846827
## class counts: 151 54
## probabilities: 0.737 0.263
## left son=3763498 (182 obs) right son=3763499 (23 obs)
## Primary splits:
## timbre_6_min < -106.1775 to the right, improve=2.3916400, (0 missing)
## timbre_4_max < 143.4215 to the right, improve=0.7451589, (0 missing)
##
## Node number 1881750: 9 observations
## predicted class=N expected loss=0.2222222 P(node) =0.001249826
## class counts: 7 2
## probabilities: 0.778 0.222
##
## Node number 1881751: 11 observations
## predicted class=Y expected loss=0.4545455 P(node) =0.001527566
## class counts: 5 6
## probabilities: 0.455 0.545
##
## Node number 3763498: 182 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2362637 P(node) =0.02527427
## class counts: 139 43
## probabilities: 0.764 0.236
## left son=7526996 (141 obs) right son=7526997 (41 obs)
## Primary splits:
## timbre_6_min < -99.4305 to the left, improve=1.1713720, (0 missing)
## timbre_4_max < 146.1955 to the left, improve=0.3386245, (0 missing)
##
## Node number 3763499: 23 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.4782609 P(node) =0.003194001
## class counts: 12 11
## probabilities: 0.522 0.478
## left son=7526998 (10 obs) right son=7526999 (13 obs)
## Primary splits:
## timbre_4_max < 153.3585 to the right, improve=1.1244150, (0 missing)
## timbre_6_min < -106.3975 to the left, improve=0.5282609, (0 missing)
## Surrogate splits:
## timbre_6_min < -106.347 to the left, agree=0.652, adj=0.2, (0 split)
##
## Node number 7526996: 141 observations
## predicted class=N expected loss=0.2056738 P(node) =0.01958061
## class counts: 112 29
## probabilities: 0.794 0.206
##
## Node number 7526997: 41 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.3414634 P(node) =0.005693654
## class counts: 27 14
## probabilities: 0.659 0.341
## left son=15053994 (33 obs) right son=15053995 (8 obs)
## Primary splits:
## timbre_6_min < -99.171 to the right, improve=1.598115, (0 missing)
## timbre_4_max < 144.1625 to the right, improve=1.598115, (0 missing)
##
## Node number 7526998: 10 observations
## predicted class=N expected loss=0.3 P(node) =0.001388696
## class counts: 7 3
## probabilities: 0.700 0.300
##
## Node number 7526999: 13 observations
## predicted class=Y expected loss=0.3846154 P(node) =0.001805305
## class counts: 5 8
## probabilities: 0.385 0.615
##
## Node number 15053994: 33 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.2727273 P(node) =0.004582697
## class counts: 24 9
## probabilities: 0.727 0.273
## left son=30107988 (8 obs) right son=30107989 (25 obs)
## Primary splits:
## timbre_6_min < -98.8505 to the left, improve=1.5709090, (0 missing)
## timbre_4_max < 169.9855 to the right, improve=0.4609091, (0 missing)
##
## Node number 15053995: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## Node number 30107988: 8 observations
## predicted class=N expected loss=0 P(node) =0.001110957
## class counts: 8 0
## probabilities: 1.000 0.000
##
## Node number 30107989: 25 observations, complexity param=0.0004716981
## predicted class=N expected loss=0.36 P(node) =0.00347174
## class counts: 16 9
## probabilities: 0.640 0.360
## left son=60215978 (17 obs) right son=60215979 (8 obs)
## Primary splits:
## timbre_6_min < -98.4525 to the right, improve=1.6523530, (0 missing)
## timbre_4_max < 154.3515 to the right, improve=0.3511688, (0 missing)
##
## Node number 60215978: 17 observations
## predicted class=N expected loss=0.2352941 P(node) =0.002360783
## class counts: 13 4
## probabilities: 0.765 0.235
##
## Node number 60215979: 8 observations
## predicted class=Y expected loss=0.375 P(node) =0.001110957
## class counts: 3 5
## probabilities: 0.375 0.625
##
## n= 7201
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 7201 1060 N (0.85279822 0.14720178)
## 2) timbre_6_min>=-86.184 4382 411 N (0.90620721 0.09379279)
## 4) timbre_4_max< 87.318 1731 67 N (0.96129405 0.03870595) *
## 5) timbre_4_max>=87.318 2651 344 N (0.87023765 0.12976235)
## 10) timbre_6_min>=-64.667 701 43 N (0.93865906 0.06134094) *
## 11) timbre_6_min< -64.667 1950 301 N (0.84564103 0.15435897)
## 22) timbre_4_max< 189.9985 1928 293 N (0.84802905 0.15197095)
## 44) timbre_6_min< -85.6325 51 2 N (0.96078431 0.03921569) *
## 45) timbre_6_min>=-85.6325 1877 291 N (0.84496537 0.15503463)
## 90) timbre_6_min>=-85.353 1847 279 N (0.84894423 0.15105577)
## 180) timbre_6_min>=-78.8265 1175 165 N (0.85957447 0.14042553)
## 360) timbre_6_min< -65.368 1118 152 N (0.86404293 0.13595707)
## 720) timbre_6_min>=-65.843 38 1 N (0.97368421 0.02631579) *
## 721) timbre_6_min< -65.843 1080 151 N (0.86018519 0.13981481)
## 1442) timbre_6_min< -68.5755 885 115 N (0.87005650 0.12994350)
## 2884) timbre_4_max< 119.0535 514 59 N (0.88521401 0.11478599) *
## 2885) timbre_4_max>=119.0535 371 56 N (0.84905660 0.15094340)
## 5770) timbre_4_max>=119.9485 360 51 N (0.85833333 0.14166667)
## 11540) timbre_6_min< -78.103 35 0 N (1.00000000 0.00000000) *
## 11541) timbre_6_min>=-78.103 325 51 N (0.84307692 0.15692308)
## 23082) timbre_6_min>=-77.8695 314 46 N (0.85350318 0.14649682)
## 46164) timbre_6_min< -76.0415 76 7 N (0.90789474 0.09210526) *
## 46165) timbre_6_min>=-76.0415 238 39 N (0.83613445 0.16386555)
## 92330) timbre_6_min>=-75.1125 204 29 N (0.85784314 0.14215686)
## 184660) timbre_4_max< 120.8475 12 0 N (1.00000000 0.00000000) *
## 184661) timbre_4_max>=120.8475 192 29 N (0.84895833 0.15104167)
## 369322) timbre_4_max>=121.6295 185 26 N (0.85945946 0.14054054)
## 738644) timbre_4_max< 123.7945 16 0 N (1.00000000 0.00000000) *
## 738645) timbre_4_max>=123.7945 169 26 N (0.84615385 0.15384615)
## 1477290) timbre_4_max>=124.9935 161 21 N (0.86956522 0.13043478) *
## 1477291) timbre_4_max< 124.9935 8 3 Y (0.37500000 0.62500000) *
## 369323) timbre_4_max< 121.6295 7 3 N (0.57142857 0.42857143) *
## 92331) timbre_6_min< -75.1125 34 10 N (0.70588235 0.29411765)
## 184662) timbre_6_min< -75.681 13 2 N (0.84615385 0.15384615) *
## 184663) timbre_6_min>=-75.681 21 8 N (0.61904762 0.38095238)
## 369326) timbre_6_min>=-75.4695 14 4 N (0.71428571 0.28571429) *
## 369327) timbre_6_min< -75.4695 7 3 Y (0.42857143 0.57142857) *
## 23083) timbre_6_min< -77.8695 11 5 N (0.54545455 0.45454545) *
## 5771) timbre_4_max< 119.9485 11 5 N (0.54545455 0.45454545) *
## 1443) timbre_6_min>=-68.5755 195 36 N (0.81538462 0.18461538)
## 2886) timbre_6_min>=-68.378 182 29 N (0.84065934 0.15934066) *
## 2887) timbre_6_min< -68.378 13 6 Y (0.46153846 0.53846154) *
## 361) timbre_6_min>=-65.368 57 13 N (0.77192982 0.22807018)
## 722) timbre_6_min>=-65.235 45 6 N (0.86666667 0.13333333) *
## 723) timbre_6_min< -65.235 12 5 Y (0.41666667 0.58333333) *
## 181) timbre_6_min< -78.8265 672 114 N (0.83035714 0.16964286)
## 362) timbre_6_min< -79.48 614 97 N (0.84201954 0.15798046)
## 724) timbre_4_max>=95.1505 551 83 N (0.84936479 0.15063521)
## 1448) timbre_6_min< -81.6945 340 44 N (0.87058824 0.12941176) *
## 1449) timbre_6_min>=-81.6945 211 39 N (0.81516588 0.18483412)
## 2898) timbre_6_min>=-80.7415 123 16 N (0.86991870 0.13008130) *
## 2899) timbre_6_min< -80.7415 88 23 N (0.73863636 0.26136364)
## 5798) timbre_6_min< -80.8265 81 19 N (0.76543210 0.23456790)
## 11596) timbre_6_min>=-81.383 58 11 N (0.81034483 0.18965517)
## 23192) timbre_6_min< -81.14 24 1 N (0.95833333 0.04166667) *
## 23193) timbre_6_min>=-81.14 34 10 N (0.70588235 0.29411765)
## 46386) timbre_6_min>=-81.1045 27 5 N (0.81481481 0.18518519) *
## 46387) timbre_6_min< -81.1045 7 2 Y (0.28571429 0.71428571) *
## 11597) timbre_6_min< -81.383 23 8 N (0.65217391 0.34782609)
## 23194) timbre_6_min< -81.452 16 4 N (0.75000000 0.25000000) *
## 23195) timbre_6_min>=-81.452 7 3 Y (0.42857143 0.57142857) *
## 5799) timbre_6_min>=-80.8265 7 3 Y (0.42857143 0.57142857) *
## 725) timbre_4_max< 95.1505 63 14 N (0.77777778 0.22222222)
## 1450) timbre_4_max< 93.8785 51 6 N (0.88235294 0.11764706) *
## 1451) timbre_4_max>=93.8785 12 4 Y (0.33333333 0.66666667) *
## 363) timbre_6_min>=-79.48 58 17 N (0.70689655 0.29310345)
## 726) timbre_6_min>=-79.4275 45 10 N (0.77777778 0.22222222)
## 1452) timbre_6_min< -79.1945 24 3 N (0.87500000 0.12500000) *
## 1453) timbre_6_min>=-79.1945 21 7 N (0.66666667 0.33333333)
## 2906) timbre_4_max>=120.2585 14 3 N (0.78571429 0.21428571) *
## 2907) timbre_4_max< 120.2585 7 3 Y (0.42857143 0.57142857) *
## 727) timbre_6_min< -79.4275 13 6 Y (0.46153846 0.53846154) *
## 91) timbre_6_min< -85.353 30 12 N (0.60000000 0.40000000)
## 182) timbre_6_min< -85.494 19 5 N (0.73684211 0.26315789) *
## 183) timbre_6_min>=-85.494 11 4 Y (0.36363636 0.63636364) *
## 23) timbre_4_max>=189.9985 22 8 N (0.63636364 0.36363636)
## 46) timbre_4_max>=204.9105 7 0 N (1.00000000 0.00000000) *
## 47) timbre_4_max< 204.9105 15 7 Y (0.46666667 0.53333333) *
## 3) timbre_6_min< -86.184 2819 649 N (0.76977652 0.23022348)
## 6) timbre_4_max< 109.353 898 132 N (0.85300668 0.14699332)
## 12) timbre_4_max< 91.4875 366 35 N (0.90437158 0.09562842)
## 24) timbre_4_max< 67.6285 92 3 N (0.96739130 0.03260870) *
## 25) timbre_4_max>=67.6285 274 32 N (0.88321168 0.11678832)
## 50) timbre_4_max>=71.4975 253 25 N (0.90118577 0.09881423)
## 100) timbre_6_min< -86.704 242 21 N (0.91322314 0.08677686)
## 200) timbre_6_min>=-96.956 138 7 N (0.94927536 0.05072464) *
## 201) timbre_6_min< -96.956 104 14 N (0.86538462 0.13461538)
## 402) timbre_4_max>=88.1385 27 0 N (1.00000000 0.00000000) *
## 403) timbre_4_max< 88.1385 77 14 N (0.81818182 0.18181818)
## 806) timbre_4_max< 84.6245 52 5 N (0.90384615 0.09615385) *
## 807) timbre_4_max>=84.6245 25 9 N (0.64000000 0.36000000)
## 1614) timbre_6_min< -103.9 12 2 N (0.83333333 0.16666667) *
## 1615) timbre_6_min>=-103.9 13 6 Y (0.46153846 0.53846154) *
## 101) timbre_6_min>=-86.704 11 4 N (0.63636364 0.36363636) *
## 51) timbre_4_max< 71.4975 21 7 N (0.66666667 0.33333333) *
## 13) timbre_4_max>=91.4875 532 97 N (0.81766917 0.18233083)
## 26) timbre_4_max>=91.831 522 91 N (0.82567050 0.17432950)
## 52) timbre_6_min>=-110.0435 434 67 N (0.84562212 0.15437788)
## 104) timbre_4_max>=105.5845 98 7 N (0.92857143 0.07142857) *
## 105) timbre_4_max< 105.5845 336 60 N (0.82142857 0.17857143)
## 210) timbre_4_max< 102.463 250 39 N (0.84400000 0.15600000) *
## 211) timbre_4_max>=102.463 86 21 N (0.75581395 0.24418605)
## 422) timbre_6_min< -98.182 26 2 N (0.92307692 0.07692308) *
## 423) timbre_6_min>=-98.182 60 19 N (0.68333333 0.31666667)
## 846) timbre_6_min>=-96.3905 52 13 N (0.75000000 0.25000000) *
## 847) timbre_6_min< -96.3905 8 2 Y (0.25000000 0.75000000) *
## 53) timbre_6_min< -110.0435 88 24 N (0.72727273 0.27272727)
## 106) timbre_4_max>=107.9985 9 0 N (1.00000000 0.00000000) *
## 107) timbre_4_max< 107.9985 79 24 N (0.69620253 0.30379747)
## 214) timbre_4_max< 104.347 53 11 N (0.79245283 0.20754717) *
## 215) timbre_4_max>=104.347 26 13 N (0.50000000 0.50000000)
## 430) timbre_6_min< -116.569 9 2 N (0.77777778 0.22222222) *
## 431) timbre_6_min>=-116.569 17 6 Y (0.35294118 0.64705882) *
## 27) timbre_4_max< 91.831 10 4 Y (0.40000000 0.60000000) *
## 7) timbre_4_max>=109.353 1921 517 N (0.73086934 0.26913066)
## 14) timbre_6_min>=-112.424 1522 380 N (0.75032852 0.24967148)
## 28) timbre_4_max>=109.674 1514 375 N (0.75231176 0.24768824)
## 56) timbre_4_max< 139.04 895 201 N (0.77541899 0.22458101)
## 112) timbre_4_max>=137.6295 31 2 N (0.93548387 0.06451613) *
## 113) timbre_4_max< 137.6295 864 199 N (0.76967593 0.23032407)
## 226) timbre_4_max< 121.5665 371 75 N (0.79784367 0.20215633)
## 452) timbre_4_max>=117.926 129 17 N (0.86821705 0.13178295) *
## 453) timbre_4_max< 117.926 242 58 N (0.76033058 0.23966942)
## 906) timbre_4_max< 116.948 217 48 N (0.77880184 0.22119816)
## 1812) timbre_4_max>=115.6655 43 4 N (0.90697674 0.09302326) *
## 1813) timbre_4_max< 115.6655 174 44 N (0.74712644 0.25287356)
## 3626) timbre_4_max< 113.117 81 15 N (0.81481481 0.18518519)
## 7252) timbre_6_min>=-89.441 11 0 N (1.00000000 0.00000000) *
## 7253) timbre_6_min< -89.441 70 15 N (0.78571429 0.21428571)
## 14506) timbre_6_min< -93.8255 46 7 N (0.84782609 0.15217391) *
## 14507) timbre_6_min>=-93.8255 24 8 N (0.66666667 0.33333333)
## 29014) timbre_4_max>=111.553 17 4 N (0.76470588 0.23529412) *
## 29015) timbre_4_max< 111.553 7 3 Y (0.42857143 0.57142857) *
## 3627) timbre_4_max>=113.117 93 29 N (0.68817204 0.31182796)
## 7254) timbre_6_min>=-102.816 66 17 N (0.74242424 0.25757576)
## 14508) timbre_6_min< -95.7075 26 3 N (0.88461538 0.11538462) *
## 14509) timbre_6_min>=-95.7075 40 14 N (0.65000000 0.35000000)
## 29018) timbre_4_max>=113.5145 30 8 N (0.73333333 0.26666667) *
## 29019) timbre_4_max< 113.5145 10 4 Y (0.40000000 0.60000000) *
## 7255) timbre_6_min< -102.816 27 12 N (0.55555556 0.44444444)
## 14510) timbre_6_min< -105.5865 17 5 N (0.70588235 0.29411765) *
## 14511) timbre_6_min>=-105.5865 10 3 Y (0.30000000 0.70000000) *
## 907) timbre_4_max>=116.948 25 10 N (0.60000000 0.40000000)
## 1814) timbre_6_min< -97.937 11 2 N (0.81818182 0.18181818) *
## 1815) timbre_6_min>=-97.937 14 6 Y (0.42857143 0.57142857) *
## 227) timbre_4_max>=121.5665 493 124 N (0.74847870 0.25152130)
## 454) timbre_4_max>=125.184 391 91 N (0.76726343 0.23273657)
## 908) timbre_4_max< 125.5545 18 1 N (0.94444444 0.05555556) *
## 909) timbre_4_max>=125.5545 373 90 N (0.75871314 0.24128686)
## 1818) timbre_4_max>=125.7355 366 86 N (0.76502732 0.23497268)
## 3636) timbre_4_max< 126.312 14 0 N (1.00000000 0.00000000) *
## 3637) timbre_4_max>=126.312 352 86 N (0.75568182 0.24431818)
## 7274) timbre_6_min>=-102.915 254 56 N (0.77952756 0.22047244)
## 14548) timbre_4_max< 134.869 195 37 N (0.81025641 0.18974359)
## 29096) timbre_4_max>=133.052 47 4 N (0.91489362 0.08510638) *
## 29097) timbre_4_max< 133.052 148 33 N (0.77702703 0.22297297)
## 58194) timbre_4_max< 129.356 73 11 N (0.84931507 0.15068493) *
## 58195) timbre_4_max>=129.356 75 22 N (0.70666667 0.29333333)
## 116390) timbre_4_max>=129.608 67 16 N (0.76119403 0.23880597)
## 232780) timbre_6_min< -95.8915 22 1 N (0.95454545 0.04545455) *
## 232781) timbre_6_min>=-95.8915 45 15 N (0.66666667 0.33333333)
## 465562) timbre_4_max< 132.4285 38 10 N (0.73684211 0.26315789)
## 931124) timbre_4_max< 130.831 12 1 N (0.91666667 0.08333333) *
## 931125) timbre_4_max>=130.831 26 9 N (0.65384615 0.34615385)
## 1862250) timbre_4_max>=131.897 13 2 N (0.84615385 0.15384615) *
## 1862251) timbre_4_max< 131.897 13 6 Y (0.46153846 0.53846154) *
## 465563) timbre_4_max>=132.4285 7 2 Y (0.28571429 0.71428571) *
## 116391) timbre_4_max< 129.608 8 2 Y (0.25000000 0.75000000) *
## 14549) timbre_4_max>=134.869 59 19 N (0.67796610 0.32203390)
## 29098) timbre_6_min>=-93.3205 28 6 N (0.78571429 0.21428571) *
## 29099) timbre_6_min< -93.3205 31 13 N (0.58064516 0.41935484)
## 58198) timbre_4_max>=135.773 19 6 N (0.68421053 0.31578947) *
## 58199) timbre_4_max< 135.773 12 5 Y (0.41666667 0.58333333) *
## 7275) timbre_6_min< -102.915 98 30 N (0.69387755 0.30612245)
## 14550) timbre_4_max>=127.5305 87 24 N (0.72413793 0.27586207)
## 29100) timbre_6_min< -108.277 24 4 N (0.83333333 0.16666667) *
## 29101) timbre_6_min>=-108.277 63 20 N (0.68253968 0.31746032)
## 58202) timbre_4_max< 130.4925 18 3 N (0.83333333 0.16666667) *
## 58203) timbre_4_max>=130.4925 45 17 N (0.62222222 0.37777778)
## 116406) timbre_4_max>=132.712 29 7 N (0.75862069 0.24137931) *
## 116407) timbre_4_max< 132.712 16 6 Y (0.37500000 0.62500000) *
## 14551) timbre_4_max< 127.5305 11 5 Y (0.45454545 0.54545455) *
## 1819) timbre_4_max< 125.7355 7 3 Y (0.42857143 0.57142857) *
## 455) timbre_4_max< 125.184 102 33 N (0.67647059 0.32352941)
## 910) timbre_6_min< -104.638 19 3 N (0.84210526 0.15789474) *
## 911) timbre_6_min>=-104.638 83 30 N (0.63855422 0.36144578)
## 1822) timbre_6_min>=-101.1935 72 23 N (0.68055556 0.31944444)
## 3644) timbre_4_max< 124.599 62 16 N (0.74193548 0.25806452)
## 7288) timbre_4_max>=123.629 21 3 N (0.85714286 0.14285714) *
## 7289) timbre_4_max< 123.629 41 13 N (0.68292683 0.31707317)
## 14578) timbre_6_min< -96.115 18 4 N (0.77777778 0.22222222) *
## 14579) timbre_6_min>=-96.115 23 9 N (0.60869565 0.39130435)
## 29158) timbre_6_min>=-93.072 14 3 N (0.78571429 0.21428571) *
## 29159) timbre_6_min< -93.072 9 3 Y (0.33333333 0.66666667) *
## 3645) timbre_4_max>=124.599 10 3 Y (0.30000000 0.70000000) *
## 1823) timbre_6_min< -101.1935 11 4 Y (0.36363636 0.63636364) *
## 57) timbre_4_max>=139.04 619 174 N (0.71890145 0.28109855)
## 114) timbre_6_min< -90.091 525 137 N (0.73904762 0.26095238)
## 228) timbre_6_min>=-91.2335 25 1 N (0.96000000 0.04000000) *
## 229) timbre_6_min< -91.2335 500 136 N (0.72800000 0.27200000)
## 458) timbre_6_min< -111.831 8 0 N (1.00000000 0.00000000) *
## 459) timbre_6_min>=-111.831 492 136 N (0.72357724 0.27642276)
## 918) timbre_6_min>=-110.8735 473 126 N (0.73361522 0.26638478)
## 1836) timbre_6_min< -110.4595 10 0 N (1.00000000 0.00000000) *
## 1837) timbre_6_min>=-110.4595 463 126 N (0.72786177 0.27213823)
## 3674) timbre_4_max>=224.1235 8 0 N (1.00000000 0.00000000) *
## 3675) timbre_4_max< 224.1235 455 126 N (0.72307692 0.27692308)
## 7350) timbre_6_min< -92.902 401 105 N (0.73815461 0.26184539)
## 14700) timbre_6_min>=-93.8975 22 2 N (0.90909091 0.09090909) *
## 14701) timbre_6_min< -93.8975 379 103 N (0.72823219 0.27176781)
## 29402) timbre_4_max< 187.1575 357 94 N (0.73669468 0.26330532)
## 58804) timbre_6_min< -96.643 288 71 N (0.75347222 0.24652778)
## 117608) timbre_6_min>=-97.0515 9 0 N (1.00000000 0.00000000) *
## 117609) timbre_6_min< -97.0515 279 71 N (0.74551971 0.25448029)
## 235218) timbre_6_min< -97.331 272 67 N (0.75367647 0.24632353)
## 470436) timbre_6_min>=-97.639 9 0 N (1.00000000 0.00000000) *
## 470437) timbre_6_min< -97.639 263 67 N (0.74524715 0.25475285)
## 940874) timbre_6_min>=-109.2095 243 59 N (0.75720165 0.24279835)
## 1881748) timbre_6_min< -107.1325 38 5 N (0.86842105 0.13157895) *
## 1881749) timbre_6_min>=-107.1325 205 54 N (0.73658537 0.26341463)
## 3763498) timbre_6_min>=-106.1775 182 43 N (0.76373626 0.23626374)
## 7526996) timbre_6_min< -99.4305 141 29 N (0.79432624 0.20567376) *
## 7526997) timbre_6_min>=-99.4305 41 14 N (0.65853659 0.34146341)
## 15053994) timbre_6_min>=-99.171 33 9 N (0.72727273 0.27272727)
## 30107988) timbre_6_min< -98.8505 8 0 N (1.00000000 0.00000000) *
## 30107989) timbre_6_min>=-98.8505 25 9 N (0.64000000 0.36000000)
## 60215978) timbre_6_min>=-98.4525 17 4 N (0.76470588 0.23529412) *
## 60215979) timbre_6_min< -98.4525 8 3 Y (0.37500000 0.62500000) *
## 15053995) timbre_6_min< -99.171 8 3 Y (0.37500000 0.62500000) *
## 3763499) timbre_6_min< -106.1775 23 11 N (0.52173913 0.47826087)
## 7526998) timbre_4_max>=153.3585 10 3 N (0.70000000 0.30000000) *
## 7526999) timbre_4_max< 153.3585 13 5 Y (0.38461538 0.61538462) *
## 940875) timbre_6_min< -109.2095 20 8 N (0.60000000 0.40000000)
## 1881750) timbre_4_max< 149.5945 9 2 N (0.77777778 0.22222222) *
## 1881751) timbre_4_max>=149.5945 11 5 Y (0.45454545 0.54545455) *
## 235219) timbre_6_min>=-97.331 7 3 Y (0.42857143 0.57142857) *
## 58805) timbre_6_min>=-96.643 69 23 N (0.66666667 0.33333333)
## 117610) timbre_6_min>=-96.346 55 13 N (0.76363636 0.23636364) *
## 117611) timbre_6_min< -96.346 14 4 Y (0.28571429 0.71428571) *
## 29403) timbre_4_max>=187.1575 22 9 N (0.59090909 0.40909091)
## 58806) timbre_4_max>=195.063 15 4 N (0.73333333 0.26666667) *
## 58807) timbre_4_max< 195.063 7 2 Y (0.28571429 0.71428571) *
## 7351) timbre_6_min>=-92.902 54 21 N (0.61111111 0.38888889)
## 14702) timbre_4_max>=141.9325 44 14 N (0.68181818 0.31818182)
## 29404) timbre_4_max< 150.4675 15 1 N (0.93333333 0.06666667) *
## 29405) timbre_4_max>=150.4675 29 13 N (0.55172414 0.44827586)
## 58810) timbre_4_max>=174.053 8 1 N (0.87500000 0.12500000) *
## 58811) timbre_4_max< 174.053 21 9 Y (0.42857143 0.57142857)
## 117622) timbre_6_min>=-92.31 12 5 N (0.58333333 0.41666667) *
## 117623) timbre_6_min< -92.31 9 2 Y (0.22222222 0.77777778) *
## 14703) timbre_4_max< 141.9325 10 3 Y (0.30000000 0.70000000) *
## 919) timbre_6_min< -110.8735 19 9 Y (0.47368421 0.52631579) *
## 115) timbre_6_min>=-90.091 94 37 N (0.60638298 0.39361702)
## 230) timbre_6_min>=-89.0365 62 19 N (0.69354839 0.30645161)
## 460) timbre_4_max>=142.605 55 14 N (0.74545455 0.25454545)
## 920) timbre_6_min< -88.4635 14 1 N (0.92857143 0.07142857) *
## 921) timbre_6_min>=-88.4635 41 13 N (0.68292683 0.31707317)
## 1842) timbre_4_max< 147.1105 10 1 N (0.90000000 0.10000000) *
## 1843) timbre_4_max>=147.1105 31 12 N (0.61290323 0.38709677)
## 3686) timbre_6_min>=-87.8225 23 7 N (0.69565217 0.30434783)
## 7372) timbre_4_max>=151.452 14 2 N (0.85714286 0.14285714) *
## 7373) timbre_4_max< 151.452 9 4 Y (0.44444444 0.55555556) *
## 3687) timbre_6_min< -87.8225 8 3 Y (0.37500000 0.62500000) *
## 461) timbre_4_max< 142.605 7 2 Y (0.28571429 0.71428571) *
## 231) timbre_6_min< -89.0365 32 14 Y (0.43750000 0.56250000)
## 462) timbre_4_max< 162.371 20 9 N (0.55000000 0.45000000)
## 924) timbre_4_max>=147.704 7 2 N (0.71428571 0.28571429) *
## 925) timbre_4_max< 147.704 13 6 Y (0.46153846 0.53846154) *
## 463) timbre_4_max>=162.371 12 3 Y (0.25000000 0.75000000) *
## 29) timbre_4_max< 109.674 8 3 Y (0.37500000 0.62500000) *
## 15) timbre_6_min< -112.424 399 137 N (0.65664160 0.34335840)
## 30) timbre_6_min< -112.758 385 129 N (0.66493506 0.33506494)
## 60) timbre_4_max< 135.5785 179 51 N (0.71508380 0.28491620)
## 120) timbre_4_max>=134.8965 9 0 N (1.00000000 0.00000000) *
## 121) timbre_4_max< 134.8965 170 51 N (0.70000000 0.30000000)
## 242) timbre_6_min< -117.385 97 24 N (0.75257732 0.24742268)
## 484) timbre_6_min>=-130.7685 75 15 N (0.80000000 0.20000000)
## 968) timbre_6_min>=-119.76 17 1 N (0.94117647 0.05882353) *
## 969) timbre_6_min< -119.76 58 14 N (0.75862069 0.24137931)
## 1938) timbre_6_min< -124.752 30 4 N (0.86666667 0.13333333) *
## 1939) timbre_6_min>=-124.752 28 10 N (0.64285714 0.35714286)
## 3878) timbre_4_max>=117.6805 21 6 N (0.71428571 0.28571429)
## 7756) timbre_6_min>=-122.317 12 1 N (0.91666667 0.08333333) *
## 7757) timbre_6_min< -122.317 9 4 Y (0.44444444 0.55555556) *
## 3879) timbre_4_max< 117.6805 7 3 Y (0.42857143 0.57142857) *
## 485) timbre_6_min< -130.7685 22 9 N (0.59090909 0.40909091)
## 970) timbre_4_max< 126.721 15 4 N (0.73333333 0.26666667) *
## 971) timbre_4_max>=126.721 7 2 Y (0.28571429 0.71428571) *
## 243) timbre_6_min>=-117.385 73 27 N (0.63013699 0.36986301)
## 486) timbre_6_min>=-114.1775 25 6 N (0.76000000 0.24000000) *
## 487) timbre_6_min< -114.1775 48 21 N (0.56250000 0.43750000)
## 974) timbre_4_max>=121.1635 28 9 N (0.67857143 0.32142857)
## 1948) timbre_4_max< 130.9325 20 4 N (0.80000000 0.20000000) *
## 1949) timbre_4_max>=130.9325 8 3 Y (0.37500000 0.62500000) *
## 975) timbre_4_max< 121.1635 20 8 Y (0.40000000 0.60000000)
## 1950) timbre_6_min< -116.3835 8 3 N (0.62500000 0.37500000) *
## 1951) timbre_6_min>=-116.3835 12 3 Y (0.25000000 0.75000000) *
## 61) timbre_4_max>=135.5785 206 78 N (0.62135922 0.37864078)
## 122) timbre_4_max>=137.1985 198 72 N (0.63636364 0.36363636)
## 244) timbre_6_min< -134.7425 23 5 N (0.78260870 0.21739130) *
## 245) timbre_6_min>=-134.7425 175 67 N (0.61714286 0.38285714)
## 490) timbre_6_min>=-120.7485 87 27 N (0.68965517 0.31034483)
## 980) timbre_4_max< 167.48 65 17 N (0.73846154 0.26153846)
## 1960) timbre_4_max>=158.734 16 1 N (0.93750000 0.06250000) *
## 1961) timbre_4_max< 158.734 49 16 N (0.67346939 0.32653061)
## 3922) timbre_4_max< 157.723 42 12 N (0.71428571 0.28571429) *
## 3923) timbre_4_max>=157.723 7 3 Y (0.42857143 0.57142857) *
## 981) timbre_4_max>=167.48 22 10 N (0.54545455 0.45454545)
## 1962) timbre_4_max>=180.2065 11 3 N (0.72727273 0.27272727) *
## 1963) timbre_4_max< 180.2065 11 4 Y (0.36363636 0.63636364) *
## 491) timbre_6_min< -120.7485 88 40 N (0.54545455 0.45454545)
## 982) timbre_6_min< -125.0005 53 21 N (0.60377358 0.39622642)
## 1964) timbre_6_min>=-130.002 32 9 N (0.71875000 0.28125000)
## 3928) timbre_4_max>=146.904 25 5 N (0.80000000 0.20000000) *
## 3929) timbre_4_max< 146.904 7 3 Y (0.42857143 0.57142857) *
## 1965) timbre_6_min< -130.002 21 9 Y (0.42857143 0.57142857)
## 3930) timbre_4_max>=152.7635 14 6 N (0.57142857 0.42857143) *
## 3931) timbre_4_max< 152.7635 7 1 Y (0.14285714 0.85714286) *
## 983) timbre_6_min>=-125.0005 35 16 Y (0.45714286 0.54285714)
## 1966) timbre_6_min>=-123.566 24 10 N (0.58333333 0.41666667)
## 3932) timbre_6_min< -121.989 12 3 N (0.75000000 0.25000000) *
## 3933) timbre_6_min>=-121.989 12 5 Y (0.41666667 0.58333333) *
## 1967) timbre_6_min< -123.566 11 2 Y (0.18181818 0.81818182) *
## 123) timbre_4_max< 137.1985 8 2 Y (0.25000000 0.75000000) *
## 31) timbre_6_min>=-112.758 14 6 Y (0.42857143 0.57142857) *
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.37657961
## 3 0.2 0.47030879
## 4 0.3 0.47032222
## 5 0.4 0.46315789
## 6 0.5 0.44954683
## 7 0.6 0.31837916
## 8 0.7 0.15709459
## 9 0.8 0.02782931
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 5848
## 2 Y 644
## Top10.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 293
## 2 416
## Prediction
## Reference N Y
## N 5848 293
## Y 644 416
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.698792e-01 3.994621e-01 8.618892e-01 8.775683e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.745099e-05 2.827705e-30
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.33207547
## 3 0.2 0.27272727
## 4 0.3 0.13043478
## 5 0.4 0.13333333
## 6 0.5 0.14117647
## 7 0.6 0.08333333
## 8 0.7 0.02985075
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 152
## 2 Y 15
## Top10.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 162
## 2 44
## Prediction
## Reference N Y
## N 152 162
## Y 15 44
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.254692e-01 1.142520e-01 4.734236e-01 5.771074e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 5.094861e-28
## model_id model_method feats
## 1 Max.cor.Y.cv.0.cp.0.rpart rpart timbre_6_min, timbre_4_max
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.559 0.113
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.7925561 0.3 0.4703222 0.8698792
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8618892 0.8775683 0.3994621 0.6506801
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.1 0.3320755 0.5254692
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4734236 0.5771074 0.114252if (glb_is_regression || glb_is_binomial) # For multinomials this model will be run next by default
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_vars,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.rpart"
## [1] " indep_vars: timbre_6_min, timbre_4_max"
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.00108 on full training set## Warning in myfit_mdl(model_id = "Max.cor.Y", model_method = "rpart",
## model_type = glb_model_type, : model's bestTune found at an extreme of
## tuneGrid for parameter: cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 7201
##
## CP nsplit rel error
## 1 0.001078167 0 1
##
## Node number 1: 7201 observations
## predicted class=N expected loss=0.1472018 P(node) =1
## class counts: 6141 1060
## probabilities: 0.853 0.147
##
## n= 7201
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 7201 1060 N (0.8527982 0.1472018) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.rpart.N
## 1 N 6141
## 2 Y 1060
## Prediction
## Reference N Y
## N 6141 0
## Y 1060 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.527982e-01 0.000000e+00 8.444027e-01 8.609070e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 5.081939e-01 4.439170e-232
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.rpart.N
## 1 N 314
## 2 Y 59
## Prediction
## Reference N Y
## N 314 0
## Y 59 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.418231e-01 0.000000e+00 8.007587e-01 8.773551e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 5.346786e-01 4.320720e-14
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.rpart rpart timbre_6_min, timbre_4_max 3
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.685 0.126 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.8328024
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8444027 0.860907 0.04053746 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8418231
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8007587 0.8773551 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.00978916 0.02058608# Used to compare vs. Interactions.High.cor.Y and/or Max.cor.Y.TmSrs
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_vars,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.glm"
## [1] " indep_vars: timbre_6_min, timbre_4_max"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.4102 -0.6047 -0.4623 -0.3333 2.6044
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.999029 0.175097 -28.550 <2e-16 ***
## timbre_6_min -0.023036 0.001835 -12.551 <2e-16 ***
## timbre_4_max 0.011122 0.001125 9.891 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 5580.7 on 7198 degrees of freedom
## AIC: 5586.7
##
## Number of Fisher Scoring iterations: 5
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.32419929
## 3 0.2 0.33236785
## 4 0.3 0.19446233
## 5 0.4 0.07582140
## 6 0.5 0.01109057
## 7 0.6 0.00000000
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.glm.N
## 1 N 4898
## 2 Y 601
## Top10.fctr.predict.Max.cor.Y.glm.Y
## 1 1243
## 2 459
## Prediction
## Reference N Y
## N 4898 1243
## Y 601 459
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.439245e-01 1.844047e-01 7.336766e-01 7.539745e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.193705e-50
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.33121019
## 3 0.2 0.33112583
## 4 0.3 0.27906977
## 5 0.4 0.06557377
## 6 0.5 0.00000000
## 7 0.6 0.00000000
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.1000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Max.cor.Y.glm.N
## 1 N 111
## 2 Y 7
## Top10.fctr.predict.Max.cor.Y.glm.Y
## 1 203
## 2 52
## Prediction
## Reference N Y
## N 111 203
## Y 7 52
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.369973e-01 9.998621e-02 3.859956e-01 4.890069e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.828054e-41
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.glm glm timbre_6_min, timbre_4_max 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.322 0.129 0.6982092
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.3323678 0.8514097
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7336766 0.7539745 0.005126623 0.7064126
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.1 0.3312102 0.4369973
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.3859956 0.4890069 0.09998621 5586.714
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001019932 0.008109304if (!is.null(glb_date_vars) &&
(sum(grepl(paste(glb_date_vars, "\\.day\\.minutes\\.poly\\.", sep=""),
names(glb_allobs_df))) > 0)) {
# ret_lst <- myfit_mdl(model_id="Max.cor.Y.TmSrs.poly1",
# model_method=ifelse(glb_is_regression, "lm",
# ifelse(glb_is_binomial, "glm", "rpart")),
# model_type=glb_model_type,
# indep_vars_vctr=c(max_cor_y_x_vars, paste0(glb_date_vars, ".day.minutes")),
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
# n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
#
ret_lst <- myfit_mdl(model_id="Max.cor.Y.TmSrs.poly",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr=c(max_cor_y_x_vars,
grep(paste(glb_date_vars, "\\.day\\.minutes\\.poly\\.", sep=""),
names(glb_allobs_df), value=TRUE)),
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
}
# Interactions.High.cor.Y
if (length(int_feats <- setdiff(unique(glb_feats_df$cor.high.X), NA)) > 0) {
# lm & glm handle interaction terms; rpart & rf do not
if (glb_is_regression || glb_is_binomial) {
indep_vars_vctr <-
c(max_cor_y_x_vars, paste(max_cor_y_x_vars[1], int_feats, sep=":"))
} else { indep_vars_vctr <- union(max_cor_y_x_vars, int_feats) }
ret_lst <- myfit_mdl(model_id="Interact.High.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
} ## [1] "fitting model: Interact.High.cor.Y.glm"
## [1] " indep_vars: timbre_6_min, timbre_4_max, timbre_6_min:timbre_0_max"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.2516 -0.5873 -0.4454 -0.3202 2.6533
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -4.9234298 0.1771093 -27.80 <2e-16 ***
## timbre_6_min -0.0720751 0.0048023 -15.01 <2e-16 ***
## timbre_4_max 0.0119132 0.0011413 10.44 <2e-16 ***
## `timbre_6_min:timbre_0_max` 0.0009463 0.0000858 11.03 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 5461.3 on 7197 degrees of freedom
## AIC: 5469.3
##
## Number of Fisher Scoring iterations: 5
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.256627527
## 2 0.1 0.336928226
## 3 0.2 0.355390335
## 4 0.3 0.257401813
## 5 0.4 0.148262548
## 6 0.5 0.076388889
## 7 0.6 0.041780200
## 8 0.7 0.009319664
## 9 0.8 0.001881468
## 10 0.9 0.000000000
## 11 1.0 0.000000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 4989
## 2 Y 582
## Top10.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 1152
## 2 478
## Prediction
## Reference N Y
## N 4989 1152
## Y 582 478
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.592001e-01 2.154280e-01 7.491514e-01 7.690385e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.658676e-42
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.35738832
## 3 0.2 0.38518519
## 4 0.3 0.27848101
## 5 0.4 0.12500000
## 6 0.5 0.06557377
## 7 0.6 0.03333333
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Interact.High.cor.Y.glm.N
## 1 N 264
## 2 Y 33
## Top10.fctr.predict.Interact.High.cor.Y.glm.Y
## 1 50
## 2 26
## Prediction
## Reference N Y
## N 264 50
## Y 33 26
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.77747989 0.25196318 0.73180986 0.81868312 0.84182306
## AccuracyPValue McnemarPValue
## 0.99955937 0.07904946
## model_id model_method
## 1 Interact.High.cor.Y.glm glm
## feats max.nTuningRuns
## 1 timbre_6_min, timbre_4_max, timbre_6_min:timbre_0_max 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.447 0.143 0.7238499
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.3553903 0.8525202
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7491514 0.7690385 0.05800737 0.7395552
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.3851852 0.7774799
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7318099 0.8186831 0.2519632 5469.265
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001478443 0.01759949# Low.cor.X
# if (glb_is_classification && glb_is_binomial)
# indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) &
# is.ConditionalX.y &
# (exclude.as.feat != 1))[, "id"] else
indep_vars_vctr <- subset(glb_feats_df, is.na(cor.high.X) & !myNearZV &
(exclude.as.feat != 1))[, "id"]
myadjust_interaction_feats <- function(vars_vctr) {
for (feat in subset(glb_feats_df, !is.na(interaction.feat))$id)
if (feat %in% vars_vctr)
vars_vctr <- union(setdiff(vars_vctr, feat),
paste0(glb_feats_df[glb_feats_df$id == feat, "interaction.feat"], ":", feat))
return(vars_vctr)
}
indep_vars_vctr <- myadjust_interaction_feats(indep_vars_vctr)
ret_lst <- myfit_mdl(model_id="Low.cor.X",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Low.cor.X.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.1030 -0.5591 -0.3601 -0.1900 3.3191
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.239e+00 7.466e-01 -2.999 0.002708 **
## timbre_4_max 6.324e-03 1.533e-03 4.125 3.70e-05 ***
## timbre_8_max 6.413e-03 2.950e-03 2.174 0.029707 *
## timbre_9_max 3.516e-03 2.377e-03 1.479 0.139078
## timbre_11_max 1.823e-02 3.342e-03 5.454 4.92e-08 ***
## timbre_6_max 3.849e-03 2.157e-03 1.785 0.074294 .
## timbre_1_max -8.007e-04 7.068e-04 -1.133 0.257241
## timbre_10_max 7.397e-03 1.731e-03 4.272 1.94e-05 ***
## timbre_2_max 4.088e-04 8.967e-04 0.456 0.648456
## tempo_confidence 5.491e-01 1.407e-01 3.901 9.57e-05 ***
## timbre_5_max 7.019e-04 7.808e-04 0.899 0.368666
## timbre_0_min 2.475e-02 4.241e-03 5.836 5.35e-09 ***
## timesignature_confidence 6.879e-01 1.923e-01 3.576 0.000348 ***
## timesignature 1.617e-01 8.739e-02 1.850 0.064313 .
## timbre_7_max -3.118e-03 1.811e-03 -1.722 0.085134 .
## key 1.736e-02 1.026e-02 1.693 0.090518 .
## .rnorm 3.092e-02 3.757e-02 0.823 0.410421
## key_confidence 2.949e-01 1.395e-01 2.115 0.034469 *
## timbre_10_min 2.963e-03 1.804e-03 1.642 0.100564
## timbre_1_min 7.136e-03 7.711e-04 9.255 < 2e-16 ***
## tempo 5.717e-04 1.665e-03 0.343 0.731358
## timbre_4_min 9.084e-03 1.953e-03 4.652 3.28e-06 ***
## timbre_3_min 6.512e-04 5.949e-04 1.095 0.273713
## timbre_3_max -2.469e-03 5.677e-04 -4.350 1.36e-05 ***
## timbre_9_min -3.823e-04 2.956e-03 -0.129 0.897106
## timbre_8_min 4.462e-03 2.810e-03 1.588 0.112321
## energy 1.830e-01 2.608e-01 0.701 0.482994
## timbre_5_min -5.646e-03 1.255e-03 -4.498 6.86e-06 ***
## pitch -5.146e+01 6.855e+00 -7.507 6.06e-14 ***
## timbre_2_min -1.564e-03 1.109e-03 -1.410 0.158465
## timbre_0_max -1.008e-01 1.178e-02 -8.557 < 2e-16 ***
## timbre_7_min -5.082e-03 1.755e-03 -2.896 0.003781 **
## timbre_11_min -2.836e-02 3.631e-03 -7.811 5.65e-15 ***
## timbre_6_min -1.612e-02 2.235e-03 -7.213 5.49e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4871.2 on 7167 degrees of freedom
## AIC: 4939.2
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.256627527
## 2 0.1 0.392010536
## 3 0.2 0.451410658
## 4 0.3 0.446480231
## 5 0.4 0.355500311
## 6 0.5 0.258684405
## 7 0.6 0.160000000
## 8 0.7 0.073542601
## 9 0.8 0.022367195
## 10 0.9 0.005644403
## 11 1.0 0.000000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Low.cor.X.glm.N
## 1 N 4978
## 2 Y 412
## Top10.fctr.predict.Low.cor.X.glm.Y
## 1 1163
## 2 648
## Prediction
## Reference N Y
## N 4978 1163
## Y 412 648
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.812804e-01 3.262998e-01 7.715504e-01 7.907822e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.179575e-79
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.45414847
## 3 0.2 0.50649351
## 4 0.3 0.49056604
## 5 0.4 0.44186047
## 6 0.5 0.30555556
## 7 0.6 0.21212121
## 8 0.7 0.12698413
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.Low.cor.X.glm.N
## 1 N 258
## 2 Y 20
## Top10.fctr.predict.Low.cor.X.glm.Y
## 1 56
## 2 39
## Prediction
## Reference N Y
## N 258 56
## Y 20 39
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.962466e-01 3.868316e-01 7.517471e-01 8.359713e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.920270e-01 5.950244e-05
## model_id model_method
## 1 Low.cor.X.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.727 0.863
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8004704 0.2 0.4514107 0.8604359
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7715504 0.7907822 0.2083714 0.841574
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.5064935 0.7962466
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7517471 0.8359713 0.3868316 4939.171
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0008623889 0.005280039rm(ret_lst)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 10 fit.models 7 0 35.689 85.439 49.75
## 11 fit.models 7 1 85.439 NA NAfit.models_1_chunk_df <- myadd_chunk(NULL, "fit.models_1_bgn")## label step_major step_minor bgn end elapsed
## 1 fit.models_1_bgn 1 0 96.739 NA NA# Options:
# 1. rpart & rf manual tuning
# 2. rf without pca (default: with pca)
#stop(here); sav_models_lst <- glb_models_lst; sav_models_df <- glb_models_df
#glb_models_lst <- sav_models_lst; glb_models_df <- sav_models_df
# All X that is not user excluded
# if (glb_is_classification && glb_is_binomial) {
# model_id_pfx <- "Conditional.X"
# # indep_vars_vctr <- setdiff(names(glb_fitobs_df), union(glb_rsp_var, glb_exclude_vars_as_features))
# indep_vars_vctr <- subset(glb_feats_df, is.ConditionalX.y &
# (exclude.as.feat != 1))[, "id"]
# } else {
model_id_pfx <- "All.X"
indep_vars_vctr <- subset(glb_feats_df, !myNearZV &
(exclude.as.feat != 1))[, "id"]
# }
indep_vars_vctr <- myadjust_interaction_feats(indep_vars_vctr)
for (method in glb_models_method_vctr) {
fit.models_1_chunk_df <- myadd_chunk(fit.models_1_chunk_df,
paste0("fit.models_1_", method), major.inc=TRUE)
if (method %in% c("rpart", "rf")) {
# rpart: fubar's the tree
# rf: skip the scenario w/ .rnorm for speed
indep_vars_vctr <- setdiff(indep_vars_vctr, c(".rnorm"))
model_id <- paste0(model_id_pfx, ".no.rnorm")
} else model_id <- model_id_pfx
ret_lst <- myfit_mdl(model_id=model_id, model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
# If All.X.glm is less accurate than Low.Cor.X.glm
# check NA coefficients & filter appropriate terms in indep_vars_vctr
# if (method == "glm") {
# orig_glm <- glb_models_lst[[paste0(model_id, ".", model_method)]]$finalModel
# orig_glm <- glb_models_lst[["All.X.glm"]]$finalModel; print(summary(orig_glm))
# vif_orig_glm <- vif(orig_glm); print(vif_orig_glm)
# print(vif_orig_glm[!is.na(vif_orig_glm) & (vif_orig_glm == Inf)])
# print(which.max(vif_orig_glm))
# print(sort(vif_orig_glm[vif_orig_glm >= 1.0e+03], decreasing=TRUE))
# glb_fitobs_df[c(1143, 3637, 3953, 4105), c("UniqueID", "Popular", "H.P.quandary", "Headline")]
# glb_feats_df[glb_feats_df$id %in% grep("[HSA]\\.nchrs.log", glb_feats_df$id, value=TRUE) | glb_feats_df$cor.high.X %in% grep("[HSA]\\.nchrs.log", glb_feats_df$id, value=TRUE), ]
# glb_feats_df[glb_feats_df$id %in% grep("[HSA]\\.npnct14.log", glb_feats_df$id, value=TRUE) | glb_feats_df$cor.high.X %in% grep("[HSA]\\.npnct14.log", glb_feats_df$id, value=TRUE), ]
# glb_feats_df[glb_feats_df$id %in% grep("[HSA]\\.T.scen", glb_feats_df$id, value=TRUE) | glb_feats_df$cor.high.X %in% grep("[HSA]\\.T.scen", glb_feats_df$id, value=TRUE), ]
# glb_feats_df[glb_feats_df$id %in% grep("[HSA]\\.P.first", glb_feats_df$id, value=TRUE) | glb_feats_df$cor.high.X %in% grep("[HSA]\\.P.first", glb_feats_df$id, value=TRUE), ]
# all.equal(glb_allobs_df$S.nuppr.log, glb_allobs_df$A.nuppr.log)
# all.equal(glb_allobs_df$S.npnct19.log, glb_allobs_df$A.npnct19.log)
# all.equal(glb_allobs_df$S.P.year.colon, glb_allobs_df$A.P.year.colon)
# all.equal(glb_allobs_df$S.T.share, glb_allobs_df$A.T.share)
# all.equal(glb_allobs_df$H.T.clip, glb_allobs_df$H.P.daily.clip.report)
# cor(glb_allobs_df$S.T.herald, glb_allobs_df$S.T.tribun)
# dsp_obs(Abstract.contains="[Dd]iar", cols=("Abstract"), all=TRUE)
# dsp_obs(Abstract.contains="[Ss]hare", cols=("Abstract"), all=TRUE)
# subset(glb_feats_df, cor.y.abs <= glb_feats_df[glb_feats_df$id == ".rnorm", "cor.y.abs"])
# corxx_mtrx <- cor(data.matrix(glb_allobs_df[, setdiff(names(glb_allobs_df), myfind_chr_cols_df(glb_allobs_df))]), use="pairwise.complete.obs"); abs_corxx_mtrx <- abs(corxx_mtrx); diag(abs_corxx_mtrx) <- 0
# which.max(abs_corxx_mtrx["S.T.tribun", ])
# abs_corxx_mtrx["A.npnct08.log", "S.npnct08.log"]
# step_glm <- step(orig_glm)
# }
# Since caret does not optimize rpart well
# if (method == "rpart")
# ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".cp.0"), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# model_type=glb_model_type,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
# n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))
}## label step_major step_minor bgn end elapsed
## 1 fit.models_1_bgn 1 0 96.739 96.754 0.015
## 2 fit.models_1_glm 2 0 96.754 NA NA
## [1] "fitting model: All.X.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9200 -0.5392 -0.3454 -0.1859 3.0866
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.471e+01 1.807e+00 8.144 3.82e-16 ***
## timbre_4_max 6.134e-03 1.551e-03 3.955 7.65e-05 ***
## timbre_8_max 4.000e-03 3.003e-03 1.332 0.182808
## timbre_9_max 1.582e-03 2.434e-03 0.650 0.515675
## timbre_11_max 1.960e-02 3.387e-03 5.787 7.16e-09 ***
## timbre_6_max 3.696e-03 2.190e-03 1.688 0.091377 .
## timbre_1_max -2.629e-04 7.156e-04 -0.367 0.713392
## timbre_10_max 5.857e-03 1.770e-03 3.310 0.000934 ***
## timbre_2_max 6.782e-04 9.068e-04 0.748 0.454477
## tempo_confidence 4.726e-01 1.422e-01 3.324 0.000887 ***
## timbre_5_max 8.640e-05 7.936e-04 0.109 0.913299
## timbre_0_min 2.310e-02 4.258e-03 5.425 5.79e-08 ***
## timesignature_confidence 7.445e-01 1.952e-01 3.814 0.000137 ***
## timesignature 1.258e-01 8.678e-02 1.450 0.147119
## timbre_7_max -3.729e-03 1.833e-03 -2.035 0.041834 *
## key 1.587e-02 1.039e-02 1.527 0.126677
## .rnorm 3.377e-02 3.808e-02 0.887 0.375170
## key_confidence 3.081e-01 1.412e-01 2.183 0.029048 *
## timbre_10_min 4.093e-03 1.839e-03 2.225 0.026075 *
## timbre_1_min 5.872e-03 7.799e-04 7.529 5.13e-14 ***
## tempo 3.876e-04 1.692e-03 0.229 0.818807
## timbre_4_min 1.036e-02 1.986e-03 5.217 1.82e-07 ***
## timbre_3_min 6.929e-04 5.985e-04 1.158 0.246945
## timbre_3_max -2.975e-03 5.818e-04 -5.114 3.15e-07 ***
## loudness 3.001e-01 2.918e-02 10.287 < 2e-16 ***
## timbre_9_min 1.413e-03 2.999e-03 0.471 0.637501
## timbre_8_min 3.880e-03 2.851e-03 1.361 0.173553
## energy -1.501e+00 3.100e-01 -4.843 1.28e-06 ***
## timbre_5_min -5.603e-03 1.277e-03 -4.388 1.15e-05 ***
## pitch -4.485e+01 6.833e+00 -6.564 5.24e-11 ***
## timbre_2_min -2.109e-03 1.126e-03 -1.872 0.061152 .
## timbre_0_max -3.313e-01 2.570e-02 -12.890 < 2e-16 ***
## timbre_7_min -4.533e-03 1.782e-03 -2.545 0.010942 *
## timbre_11_min -2.625e-02 3.694e-03 -7.105 1.20e-12 ***
## timbre_6_min -1.685e-02 2.264e-03 -7.444 9.77e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4758.4 on 7166 degrees of freedom
## AIC: 4828.4
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.40895591
## 3 0.2 0.46669017
## 4 0.3 0.45406949
## 5 0.4 0.39522388
## 6 0.5 0.28981349
## 7 0.6 0.19614148
## 8 0.7 0.12780656
## 9 0.8 0.05469462
## 10 0.9 0.01683817
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.All.X.glm.N Top10.fctr.predict.All.X.glm.Y
## 1 N 5026 1115
## 2 Y 398 662
## Prediction
## Reference N Y
## N 5026 1115
## Y 398 662
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.898903e-01 3.461100e-01 7.802951e-01 7.992502e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.144941e-75
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.42975207
## 3 0.2 0.52229299
## 4 0.3 0.47706422
## 5 0.4 0.44705882
## 6 0.5 0.32432432
## 7 0.6 0.23880597
## 8 0.7 0.15625000
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.All.X.glm.N Top10.fctr.predict.All.X.glm.Y
## 1 N 257 57
## 2 Y 18 41
## Prediction
## Reference N Y
## N 257 57
## Y 18 41
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.989276e-01 4.047492e-01 7.546059e-01 8.384302e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.886242e-01 1.144703e-05
## model_id model_method
## 1 All.X.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 5.848 0.886
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8139835 0.2 0.4666902 0.860436
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7802951 0.7992502 0.2286063 0.8421138
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.522293 0.7989276
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7546059 0.8384302 0.4047492 4828.367
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001667005 0.01770052
## label step_major step_minor bgn end elapsed
## 2 fit.models_1_glm 2 0 96.754 109.988 13.234
## 3 fit.models_1_bayesglm 3 0 109.989 NA NA
## [1] "fitting model: All.X.bayesglm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"## Loading required package: arm
## Loading required package: MASS
##
## Attaching package: 'MASS'
##
## The following object is masked from 'package:dplyr':
##
## select
##
## Loading required package: Matrix
## Loading required package: lme4
##
## arm (Version 1.8-5, built: 2015-05-13)
##
## Working directory is /Users/bbalaji-2012/Documents/Work/Courses/MIT/Analytics_Edge_15_071x/Assignments/HW3_Billboard_Songs
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9147 -0.5393 -0.3464 -0.1866 3.0803
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.439e+01 1.790e+00 8.041 8.88e-16 ***
## timbre_4_max 6.128e-03 1.547e-03 3.963 7.41e-05 ***
## timbre_8_max 4.059e-03 2.996e-03 1.355 0.175449
## timbre_9_max 1.625e-03 2.427e-03 0.670 0.502949
## timbre_11_max 1.951e-02 3.380e-03 5.771 7.86e-09 ***
## timbre_6_max 3.710e-03 2.184e-03 1.698 0.089428 .
## timbre_1_max -2.664e-04 7.139e-04 -0.373 0.709069
## timbre_10_max 5.864e-03 1.766e-03 3.321 0.000898 ***
## timbre_2_max 6.697e-04 9.045e-04 0.740 0.459054
## tempo_confidence 4.733e-01 1.419e-01 3.336 0.000850 ***
## timbre_5_max 9.650e-05 7.916e-04 0.122 0.902972
## timbre_0_min 2.309e-02 4.251e-03 5.430 5.62e-08 ***
## timesignature_confidence 7.408e-01 1.947e-01 3.805 0.000142 ***
## timesignature 1.262e-01 8.659e-02 1.457 0.145150
## timbre_7_max -3.718e-03 1.828e-03 -2.034 0.041942 *
## key 1.585e-02 1.037e-02 1.528 0.126484
## .rnorm 3.367e-02 3.802e-02 0.885 0.375896
## key_confidence 3.067e-01 1.409e-01 2.177 0.029511 *
## timbre_10_min 4.078e-03 1.833e-03 2.225 0.026071 *
## timbre_1_min 5.860e-03 7.774e-04 7.538 4.78e-14 ***
## tempo 3.900e-04 1.688e-03 0.231 0.817295
## timbre_4_min 1.030e-02 1.981e-03 5.199 2.01e-07 ***
## timbre_3_min 6.921e-04 5.971e-04 1.159 0.246444
## timbre_3_max -2.954e-03 5.792e-04 -5.100 3.40e-07 ***
## loudness 2.947e-01 2.887e-02 10.210 < 2e-16 ***
## timbre_9_min 1.399e-03 2.991e-03 0.468 0.639990
## timbre_8_min 3.858e-03 2.844e-03 1.357 0.174909
## energy -1.474e+00 3.081e-01 -4.783 1.73e-06 ***
## timbre_5_min -5.583e-03 1.273e-03 -4.385 1.16e-05 ***
## pitch -4.456e+01 6.785e+00 -6.568 5.12e-11 ***
## timbre_2_min -2.103e-03 1.123e-03 -1.874 0.060997 .
## timbre_0_max -3.267e-01 2.545e-02 -12.838 < 2e-16 ***
## timbre_7_min -4.527e-03 1.777e-03 -2.548 0.010846 *
## timbre_11_min -2.622e-02 3.686e-03 -7.112 1.15e-12 ***
## timbre_6_min -1.685e-02 2.262e-03 -7.450 9.34e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4758.4 on 7166 degrees of freedom
## AIC: 4828.4
##
## Number of Fisher Scoring iterations: 7
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.40866591
## 3 0.2 0.46506704
## 4 0.3 0.45523810
## 5 0.4 0.39450090
## 6 0.5 0.28879310
## 7 0.6 0.19516129
## 8 0.7 0.12780656
## 9 0.8 0.05296804
## 10 0.9 0.01683817
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.All.X.bayesglm.N
## 1 N 5026
## 2 Y 401
## Top10.fctr.predict.All.X.bayesglm.Y
## 1 1115
## 2 659
## Prediction
## Reference N Y
## N 5026 1115
## Y 401 659
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.894737e-01 3.442135e-01 7.798718e-01 7.988406e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 6.617306e-75
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.42975207
## 3 0.2 0.52229299
## 4 0.3 0.47706422
## 5 0.4 0.44705882
## 6 0.5 0.30136986
## 7 0.6 0.23880597
## 8 0.7 0.15625000
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.All.X.bayesglm.N
## 1 N 257
## 2 Y 18
## Top10.fctr.predict.All.X.bayesglm.Y
## 1 57
## 2 41
## Prediction
## Reference N Y
## N 257 57
## Y 18 41
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.989276e-01 4.047492e-01 7.546059e-01 8.384302e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.886242e-01 1.144703e-05
## model_id model_method
## 1 All.X.bayesglm bayesglm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 4.142 1.068
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8139566 0.2 0.465067 0.860575
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7798718 0.7988406 0.2256497 0.8423297
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.522293 0.7989276
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7546059 0.8384302 0.4047492 4828.409
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001725688 0.01758228
## label step_major step_minor bgn end elapsed
## 3 fit.models_1_bayesglm 3 0 109.989 119.852 9.863
## 4 fit.models_1_rpart 4 0 119.853 NA NA
## [1] "fitting model: All.X.no.rnorm.rpart"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.0217 on full training set## Warning in myfit_mdl(model_id = model_id, model_method = method,
## indep_vars_vctr = indep_vars_vctr, : model's bestTune found at an extreme
## of tuneGrid for parameter: cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 7201
##
## CP nsplit rel error
## 1 0.02169811 0 1
##
## Node number 1: 7201 observations
## predicted class=N expected loss=0.1472018 P(node) =1
## class counts: 6141 1060
## probabilities: 0.853 0.147
##
## n= 7201
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 7201 1060 N (0.8527982 0.1472018) *
## [1] " calling mypredict_mdl for fit:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.All.X.no.rnorm.rpart.N
## 1 N 6141
## 2 Y 1060
## Prediction
## Reference N Y
## N 6141 0
## Y 1060 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.527982e-01 0.000000e+00 8.444027e-01 8.609070e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 5.081939e-01 4.439170e-232
## [1] " calling mypredict_mdl for OOB:"
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.All.X.no.rnorm.rpart.N
## 1 N 314
## 2 Y 59
## Prediction
## Reference N Y
## N 314 0
## Y 59 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.418231e-01 0.000000e+00 8.007587e-01 8.773551e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 5.346786e-01 4.320720e-14
## model_id model_method
## 1 All.X.no.rnorm.rpart rpart
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 4.588 0.865
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0.5 0 0.8527983
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8444027 0.860907 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.8418231
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.8007587 0.8773551 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0002050944 0
## label step_major step_minor bgn end elapsed
## 4 fit.models_1_rpart 4 0 119.853 126.233 6.38
## 5 fit.models_1_rf 5 0 126.234 NA NA
## [1] "fitting model: All.X.no.rnorm.rf"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"## Loading required package: randomForest
## randomForest 4.6-10
## Type rfNews() to see new features/changes/bug fixes.
##
## Attaching package: 'randomForest'
##
## The following object is masked from 'package:dplyr':
##
## combine
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 17 on full training set
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 7201 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 14402 matrix numeric
## oob.times 7201 -none- numeric
## classes 2 -none- character
## importance 33 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 7201 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 33 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.2566275
## 2 0.1 0.7282721
## 3 0.2 0.9774089
## 4 0.3 0.9971778
## 5 0.4 0.9990575
## 6 0.5 0.9990557
## 7 0.6 0.9990557
## 8 0.7 0.8248337
## 9 0.8 0.3161239
## 10 0.9 0.0674567
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 6139
## 2 Y NA
## Top10.fctr.predict.All.X.no.rnorm.rf.Y
## 1 2
## 2 1060
## Prediction
## Reference N Y
## N 6139 2
## Y 0 1060
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.9997223 0.9988946 0.9989971 0.9999664 0.8527982
## AccuracyPValue McnemarPValue
## 0.0000000 0.4795001
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.37113402
## 3 0.2 0.47674419
## 4 0.3 0.46464646
## 5 0.4 0.32876712
## 6 0.5 0.20895522
## 7 0.6 0.15625000
## 8 0.7 0.03333333
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 242
## 2 Y 18
## Top10.fctr.predict.All.X.no.rnorm.rf.Y
## 1 72
## 2 41
## Prediction
## Reference N Y
## N 242 72
## Y 18 41
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.587131e-01 3.394593e-01 7.119951e-01 8.012709e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.999879e-01 2.314376e-08
## model_id model_method
## 1 All.X.no.rnorm.rf rf
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 76.855 25.191
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9999997 0.4 0.9990575 0.8647411
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9989971 0.9999664 0.1813316 0.8103206
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.4767442 0.7587131
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7119951 0.8012709 0.3394593
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.001022467 0.01236912# User specified
# Ensure at least 2 vars in each regression; else varImp crashes
# sav_models_lst <- glb_models_lst; sav_models_df <- glb_models_df; sav_featsimp_df <- glb_featsimp_df
# glb_models_lst <- sav_models_lst; glb_models_df <- sav_models_df; glm_featsimp_df <- sav_featsimp_df
# easier to exclude features
require(gdata) # needed for trim## Loading required package: gdata
## gdata: read.xls support for 'XLS' (Excel 97-2004) files ENABLED.
##
## gdata: read.xls support for 'XLSX' (Excel 2007+) files ENABLED.
##
## Attaching package: 'gdata'
##
## The following object is masked from 'package:randomForest':
##
## combine
##
## The following objects are masked from 'package:dplyr':
##
## combine, first, last
##
## The following object is masked from 'package:stats':
##
## nobs
##
## The following object is masked from 'package:utils':
##
## object.sizemodel_id <- "SongsLog1";
indep_vars_vctr <- head(subset(glb_models_df, grepl("All\\.X\\.", model_id), select=feats)
, 1)[, "feats"]
indep_vars_vctr <- trim(unlist(strsplit(indep_vars_vctr, "[,]")))
indep_vars_vctr <- setdiff(indep_vars_vctr, ".rnorm")
# easier to include features
#model_id <- "Rank9.2"; indep_vars_vctr <- c(NULL
# ,"<feat1>"
# ,"<feat1>*<feat2>"
# ,"<feat1>:<feat2>"
# )
for (method in c("glm")) {
ret_lst <- myfit_mdl(model_id=model_id, model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
csm_mdl_id <- paste0(model_id, ".", method)
csm_featsimp_df <- myget_feats_importance(glb_models_lst[[paste0(model_id, ".", method)]]); print(head(csm_featsimp_df))
}## [1] "fitting model: SongsLog1.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9220 -0.5399 -0.3459 -0.1845 3.0770
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.470e+01 1.806e+00 8.138 4.03e-16 ***
## timbre_4_max 6.110e-03 1.550e-03 3.942 8.10e-05 ***
## timbre_8_max 4.011e-03 3.003e-03 1.336 0.181620
## timbre_9_max 1.603e-03 2.434e-03 0.659 0.510188
## timbre_11_max 1.967e-02 3.385e-03 5.811 6.21e-09 ***
## timbre_6_max 3.668e-03 2.190e-03 1.675 0.093875 .
## timbre_1_max -2.449e-04 7.152e-04 -0.342 0.732087
## timbre_10_max 5.825e-03 1.769e-03 3.292 0.000995 ***
## timbre_2_max 6.586e-04 9.066e-04 0.726 0.467571
## tempo_confidence 4.732e-01 1.422e-01 3.329 0.000873 ***
## timbre_5_max 7.736e-05 7.935e-04 0.097 0.922337
## timbre_0_min 2.316e-02 4.256e-03 5.441 5.29e-08 ***
## timesignature_confidence 7.450e-01 1.953e-01 3.815 0.000136 ***
## timesignature 1.264e-01 8.674e-02 1.457 0.145050
## timbre_7_max -3.774e-03 1.832e-03 -2.060 0.039408 *
## key 1.588e-02 1.039e-02 1.529 0.126349
## key_confidence 3.087e-01 1.412e-01 2.187 0.028760 *
## timbre_10_min 4.126e-03 1.839e-03 2.244 0.024852 *
## timbre_1_min 5.881e-03 7.798e-04 7.542 4.64e-14 ***
## tempo 3.634e-04 1.691e-03 0.215 0.829889
## timbre_4_min 1.040e-02 1.985e-03 5.237 1.63e-07 ***
## timbre_3_min 6.920e-04 5.985e-04 1.156 0.247583
## timbre_3_max -2.967e-03 5.815e-04 -5.103 3.34e-07 ***
## loudness 2.999e-01 2.917e-02 10.282 < 2e-16 ***
## timbre_9_min 1.367e-03 2.998e-03 0.456 0.648356
## timbre_8_min 3.911e-03 2.851e-03 1.372 0.170123
## energy -1.502e+00 3.099e-01 -4.847 1.25e-06 ***
## timbre_5_min -5.598e-03 1.277e-03 -4.385 1.16e-05 ***
## pitch -4.491e+01 6.835e+00 -6.570 5.02e-11 ***
## timbre_2_min -2.127e-03 1.126e-03 -1.889 0.058843 .
## timbre_0_max -3.310e-01 2.569e-02 -12.882 < 2e-16 ***
## timbre_7_min -4.549e-03 1.781e-03 -2.554 0.010661 *
## timbre_11_min -2.625e-02 3.693e-03 -7.108 1.18e-12 ***
## timbre_6_min -1.686e-02 2.264e-03 -7.445 9.66e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4759.2 on 7167 degrees of freedom
## AIC: 4827.2
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.41122775
## 3 0.2 0.47046339
## 4 0.3 0.45700713
## 5 0.4 0.39689181
## 6 0.5 0.28939828
## 7 0.6 0.19386107
## 8 0.7 0.12769629
## 9 0.8 0.05818182
## 10 0.9 0.01683817
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.SongsLog1.glm.N
## 1 N 5039
## 2 Y 395
## Top10.fctr.predict.SongsLog1.glm.Y
## 1 1102
## 2 665
## Prediction
## Reference N Y
## N 5039 1102
## Y 395 665
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.921122e-01 3.510460e-01 7.825528e-01 8.014345e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.181971e-74
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.44628099
## 3 0.2 0.51612903
## 4 0.3 0.49090909
## 5 0.4 0.43678161
## 6 0.5 0.30136986
## 7 0.6 0.23880597
## 8 0.7 0.15625000
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.SongsLog1.glm.N
## 1 N 258
## 2 Y 19
## Top10.fctr.predict.SongsLog1.glm.Y
## 1 56
## 2 40
## Prediction
## Reference N Y
## N 258 56
## Y 19 40
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.989276e-01 3.982189e-01 7.546059e-01 8.384302e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.886242e-01 3.225641e-05
## model_id model_method
## 1 SongsLog1.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 3.527 0.827
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8138747 0.2 0.4704634 0.8609916
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7825528 0.8014345 0.2284371 0.8425996
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.516129 0.7989276
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7546059 0.8384302 0.3982189 4827.154
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.002130866 0.01576222
## importance
## timbre_0_max 100.00000
## loudness 79.66080
## timbre_1_min 58.22552
## timbre_6_min 57.47385
## timbre_11_min 54.83557
## pitch 50.62914model_id <- "SongsLog2"; indep_vars_vctr <- setdiff(indep_vars_vctr, "loudness")
for (method in c("glm")) {
ret_lst <- myfit_mdl(model_id=model_id, model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
csm_mdl_id <- paste0(model_id, ".", method)
csm_featsimp_df <- myget_feats_importance(glb_models_lst[[paste0(model_id, ".", method)]]); print(head(csm_featsimp_df))
}## [1] "fitting model: SongsLog2.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.0983 -0.5607 -0.3602 -0.1902 3.3107
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.241e+00 7.465e-01 -3.002 0.002686 **
## timbre_4_max 6.306e-03 1.532e-03 4.115 3.87e-05 ***
## timbre_8_max 6.423e-03 2.950e-03 2.177 0.029497 *
## timbre_9_max 3.525e-03 2.377e-03 1.483 0.138017
## timbre_11_max 1.829e-02 3.341e-03 5.476 4.34e-08 ***
## timbre_6_max 3.814e-03 2.157e-03 1.768 0.076982 .
## timbre_1_max -7.830e-04 7.064e-04 -1.108 0.267650
## timbre_10_max 7.367e-03 1.731e-03 4.255 2.09e-05 ***
## timbre_2_max 3.889e-04 8.964e-04 0.434 0.664427
## tempo_confidence 5.497e-01 1.407e-01 3.906 9.40e-05 ***
## timbre_5_max 6.937e-04 7.807e-04 0.889 0.374256
## timbre_0_min 2.479e-02 4.240e-03 5.847 5.01e-09 ***
## timesignature_confidence 6.885e-01 1.924e-01 3.578 0.000346 ***
## timesignature 1.625e-01 8.734e-02 1.860 0.062873 .
## timbre_7_max -3.158e-03 1.811e-03 -1.744 0.081090 .
## key 1.740e-02 1.026e-02 1.697 0.089740 .
## key_confidence 2.954e-01 1.394e-01 2.118 0.034163 *
## timbre_10_min 2.993e-03 1.804e-03 1.660 0.097004 .
## timbre_1_min 7.143e-03 7.710e-04 9.265 < 2e-16 ***
## tempo 5.521e-04 1.665e-03 0.332 0.740226
## timbre_4_min 9.115e-03 1.952e-03 4.670 3.02e-06 ***
## timbre_3_min 6.500e-04 5.949e-04 1.093 0.274524
## timbre_3_max -2.462e-03 5.674e-04 -4.339 1.43e-05 ***
## timbre_9_min -4.282e-04 2.955e-03 -0.145 0.884792
## timbre_8_min 4.488e-03 2.810e-03 1.597 0.110254
## energy 1.813e-01 2.608e-01 0.695 0.486991
## timbre_5_min -5.641e-03 1.255e-03 -4.495 6.95e-06 ***
## pitch -5.150e+01 6.857e+00 -7.511 5.87e-14 ***
## timbre_2_min -1.579e-03 1.109e-03 -1.424 0.154531
## timbre_0_max -1.007e-01 1.178e-02 -8.551 < 2e-16 ***
## timbre_7_min -5.102e-03 1.755e-03 -2.907 0.003644 **
## timbre_11_min -2.837e-02 3.630e-03 -7.815 5.48e-15 ***
## timbre_6_min -1.612e-02 2.235e-03 -7.214 5.45e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4871.8 on 7168 degrees of freedom
## AIC: 4937.8
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.256627527
## 2 0.1 0.392096597
## 3 0.2 0.450871080
## 4 0.3 0.448192771
## 5 0.4 0.357986327
## 6 0.5 0.258112094
## 7 0.6 0.164588529
## 8 0.7 0.075201432
## 9 0.8 0.024186047
## 10 0.9 0.005644403
## 11 1.0 0.000000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.SongsLog2.glm.N
## 1 N 4978
## 2 Y 413
## Top10.fctr.predict.SongsLog2.glm.Y
## 1 1163
## 2 647
## Prediction
## Reference N Y
## N 4978 1163
## Y 413 647
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.811415e-01 3.256685e-01 7.714094e-01 7.906455e-01 8.527982e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.129046e-79
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.45814978
## 3 0.2 0.51612903
## 4 0.3 0.49056604
## 5 0.4 0.44186047
## 6 0.5 0.30555556
## 7 0.6 0.21212121
## 8 0.7 0.12698413
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.SongsLog2.glm.N
## 1 N 258
## 2 Y 19
## Top10.fctr.predict.SongsLog2.glm.Y
## 1 56
## 2 40
## Prediction
## Reference N Y
## N 258 56
## Y 19 40
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.989276e-01 3.982189e-01 7.546059e-01 8.384302e-01 8.418231e-01
## AccuracyPValue McnemarPValue
## 9.886242e-01 3.225641e-05
## model_id model_method
## 1 SongsLog2.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.575 0.803
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8003625 0.2 0.4508711 0.8601582
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7714094 0.7906455 0.2067826 0.8430854
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.516129 0.7989276
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.7546059 0.8384302 0.3982189 4937.849
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0006681795 0.005733383
## importance
## timbre_1_min 100.00000
## timbre_0_max 92.16662
## timbre_11_min 84.10097
## pitch 80.76217
## timbre_6_min 77.50214
## timbre_0_min 62.51903model_id <- "SongsLog3";
indep_vars_vctr <- head(subset(glb_models_df, grepl("All\\.X\\.", model_id), select=feats)
, 1)[, "feats"]
indep_vars_vctr <- trim(unlist(strsplit(indep_vars_vctr, "[,]")))
indep_vars_vctr <- setdiff(indep_vars_vctr, c(".rnorm", "energy"))
for (method in c("glm")) {
ret_lst <- myfit_mdl(model_id=model_id, model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
csm_mdl_id <- paste0(model_id, ".", method)
csm_featsimp_df <- myget_feats_importance(glb_models_lst[[paste0(model_id, ".", method)]]); print(head(csm_featsimp_df))
}## [1] "fitting model: SongsLog3.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9182 -0.5417 -0.3481 -0.1874 3.4171
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.196e+01 1.714e+00 6.977 3.01e-12 ***
## timbre_4_max 6.467e-03 1.541e-03 4.196 2.72e-05 ***
## timbre_8_max 4.658e-03 2.988e-03 1.559 0.119022
## timbre_9_max 1.342e-03 2.424e-03 0.554 0.579900
## timbre_11_max 1.984e-02 3.365e-03 5.896 3.74e-09 ***
## timbre_6_max 3.447e-03 2.182e-03 1.580 0.114203
## timbre_1_max -5.115e-04 7.110e-04 -0.719 0.471928
## timbre_10_max 5.793e-03 1.759e-03 3.294 0.000988 ***
## timbre_2_max 4.119e-04 9.020e-04 0.457 0.647915
## tempo_confidence 3.841e-01 1.398e-01 2.747 0.006019 **
## timbre_5_max 2.979e-04 7.855e-04 0.379 0.704526
## timbre_0_min 2.205e-02 4.239e-03 5.200 1.99e-07 ***
## timesignature_confidence 7.143e-01 1.946e-01 3.670 0.000242 ***
## timesignature 1.151e-01 8.726e-02 1.319 0.187183
## timbre_7_max -3.394e-03 1.820e-03 -1.865 0.062208 .
## key 1.649e-02 1.035e-02 1.593 0.111056
## key_confidence 3.394e-01 1.409e-01 2.409 0.015984 *
## timbre_10_min 4.050e-03 1.827e-03 2.217 0.026637 *
## timbre_1_min 5.416e-03 7.643e-04 7.086 1.38e-12 ***
## tempo -6.460e-04 1.665e-03 -0.388 0.698107
## timbre_4_min 1.105e-02 1.978e-03 5.585 2.34e-08 ***
## timbre_3_min 3.179e-04 5.869e-04 0.542 0.588083
## timbre_3_max -2.964e-03 5.758e-04 -5.147 2.64e-07 ***
## loudness 2.306e-01 2.528e-02 9.120 < 2e-16 ***
## timbre_9_min -9.318e-05 2.957e-03 -0.032 0.974859
## timbre_8_min 3.686e-03 2.833e-03 1.301 0.193229
## timbre_5_min -5.135e-03 1.269e-03 -4.046 5.21e-05 ***
## pitch -5.328e+01 6.733e+00 -7.914 2.49e-15 ***
## timbre_2_min -2.254e-03 1.120e-03 -2.012 0.044190 *
## timbre_0_max -3.105e-01 2.537e-02 -12.240 < 2e-16 ***
## timbre_7_min -5.128e-03 1.768e-03 -2.900 0.003733 **
## timbre_11_min -2.638e-02 3.683e-03 -7.162 7.96e-13 ***
## timbre_6_min -1.784e-02 2.246e-03 -7.945 1.94e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4782.7 on 7168 degrees of freedom
## AIC: 4848.7
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.40791269
## 3 0.2 0.45992300
## 4 0.3 0.46445498
## 5 0.4 0.38008500
## 6 0.5 0.28034682
## 7 0.6 0.18800648
## 8 0.7 0.12000000
## 9 0.8 0.04044118
## 10 0.9 0.01312090
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.SongsLog3.glm.N
## 1 N 5581
## 2 Y 570
## Top10.fctr.predict.SongsLog3.glm.Y
## 1 560
## 2 490
## Prediction
## Reference N Y
## N 5581 560
## Y 570 490
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8430774 0.3725277 0.8344684 0.8514076 0.8527982
## AccuracyPValue McnemarPValue
## 0.9900387 0.7889042
## [1] " calling mypredict_mdl for OOB:"
## threshold f.score
## 1 0.0 0.27314815
## 2 0.1 0.45188285
## 3 0.2 0.49350649
## 4 0.3 0.50877193
## 5 0.4 0.47191011
## 6 0.5 0.35135135
## 7 0.6 0.21212121
## 8 0.7 0.15625000
## 9 0.8 0.03333333
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## Top10.fctr Top10.fctr.predict.SongsLog3.glm.N
## 1 N 288
## 2 Y 30
## Top10.fctr.predict.SongsLog3.glm.Y
## 1 26
## 2 29
## Prediction
## Reference N Y
## N 288 26
## Y 30 29
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8498660 0.4202931 0.8095092 0.8845549 0.8418231
## AccuracyPValue McnemarPValue
## 0.3667675 0.6884997
## model_id model_method
## 1 SongsLog3.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.638 0.812
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8115728 0.3 0.464455 0.860575
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8344684 0.8514076 0.2189265 0.848969
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.5087719 0.849866
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.8095092 0.8845549 0.4202931 4848.719
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.000611418 0.01842614
## importance
## timbre_0_max 100.00000
## loudness 74.44375
## timbre_6_min 64.81746
## pitch 64.56445
## timbre_11_min 58.40369
## timbre_1_min 57.78607# Ntv.1.lm <- lm(reformulate(indep_vars_vctr, glb_rsp_var), glb_trnobs_df); print(summary(Ntv.1.lm))
#print(dsp_models_df <- orderBy(model_sel_frmla, glb_models_df)[, dsp_models_cols])
#csm_featsimp_df[grepl("H.npnct19.log", row.names(csm_featsimp_df)), , FALSE]
#csm_OOBobs_df <- glb_get_predictions(glb_OOBobs_df, mdl_id=csm_mdl_id, rsp_var_out=glb_rsp_var_out, prob_threshold_def=glb_models_df[glb_models_df$model_id == csm_mdl_id, "opt.prob.threshold.OOB"])
#print(sprintf("%s OOB confusion matrix & accuracy: ", csm_mdl_id)); print(t(confusionMatrix(csm_OOBobs_df[, paste0(glb_rsp_var_out, csm_mdl_id)], csm_OOBobs_df[, glb_rsp_var])$table))
#glb_models_df[, "max.Accuracy.OOB", FALSE]
#varImp(glb_models_lst[["Low.cor.X.glm"]])
#orderBy(~ -Overall, varImp(glb_models_lst[["All.X.2.glm"]])$importance)
#orderBy(~ -Overall, varImp(glb_models_lst[["All.X.3.glm"]])$importance)
#glb_feats_df[grepl("npnct28", glb_feats_df$id), ]
#print(sprintf("%s OOB confusion matrix & accuracy: ", glb_sel_mdl_id)); print(t(confusionMatrix(glb_OOBobs_df[, paste0(glb_rsp_var_out, glb_sel_mdl_id)], glb_OOBobs_df[, glb_rsp_var])$table))
# User specified bivariate models
# indep_vars_vctr_lst <- list()
# for (feat in setdiff(names(glb_fitobs_df),
# union(glb_rsp_var, glb_exclude_vars_as_features)))
# indep_vars_vctr_lst[["feat"]] <- feat
# User specified combinatorial models
# indep_vars_vctr_lst <- list()
# combn_mtrx <- combn(c("<feat1_name>", "<feat2_name>", "<featn_name>"),
# <num_feats_to_choose>)
# for (combn_ix in 1:ncol(combn_mtrx))
# #print(combn_mtrx[, combn_ix])
# indep_vars_vctr_lst[[combn_ix]] <- combn_mtrx[, combn_ix]
# template for myfit_mdl
# rf is hard-coded in caret to recognize only Accuracy / Kappa evaluation metrics
# only for OOB in trainControl ?
# ret_lst <- myfit_mdl_fn(model_id=paste0(model_id_pfx, ""), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_fitobs_df, OOB_df=glb_OOBobs_df,
# n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df,
# model_loss_mtrx=glb_model_metric_terms,
# model_summaryFunction=glb_model_metric_smmry,
# model_metric=glb_model_metric,
# model_metric_maximize=glb_model_metric_maximize)
# Simplify a model
# fit_df <- glb_fitobs_df; glb_mdl <- step(<complex>_mdl)
# Non-caret models
# rpart_area_mdl <- rpart(reformulate("Area", response=glb_rsp_var),
# data=glb_fitobs_df, #method="class",
# control=rpart.control(cp=0.12),
# parms=list(loss=glb_model_metric_terms))
# print("rpart_sel_wlm_mdl"); prp(rpart_sel_wlm_mdl)
#
print(glb_models_df)## model_id model_method
## MFO.myMFO_classfr MFO.myMFO_classfr myMFO_classfr
## Random.myrandom_classfr Random.myrandom_classfr myrandom_classfr
## Max.cor.Y.cv.0.rpart Max.cor.Y.cv.0.rpart rpart
## Max.cor.Y.cv.0.cp.0.rpart Max.cor.Y.cv.0.cp.0.rpart rpart
## Max.cor.Y.rpart Max.cor.Y.rpart rpart
## Max.cor.Y.glm Max.cor.Y.glm glm
## Interact.High.cor.Y.glm Interact.High.cor.Y.glm glm
## Low.cor.X.glm Low.cor.X.glm glm
## All.X.glm All.X.glm glm
## All.X.bayesglm All.X.bayesglm bayesglm
## All.X.no.rnorm.rpart All.X.no.rnorm.rpart rpart
## All.X.no.rnorm.rf All.X.no.rnorm.rf rf
## SongsLog1.glm SongsLog1.glm glm
## SongsLog2.glm SongsLog2.glm glm
## SongsLog3.glm SongsLog3.glm glm
## feats
## MFO.myMFO_classfr .rnorm
## Random.myrandom_classfr .rnorm
## Max.cor.Y.cv.0.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.cv.0.cp.0.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.glm timbre_6_min, timbre_4_max
## Interact.High.cor.Y.glm timbre_6_min, timbre_4_max, timbre_6_min:timbre_0_max
## Low.cor.X.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.bayesglm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.no.rnorm.rpart timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.no.rnorm.rf timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog1.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog2.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog3.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything
## MFO.myMFO_classfr 0 0.318
## Random.myrandom_classfr 0 0.235
## Max.cor.Y.cv.0.rpart 0 0.843
## Max.cor.Y.cv.0.cp.0.rpart 0 0.559
## Max.cor.Y.rpart 3 1.685
## Max.cor.Y.glm 1 1.322
## Interact.High.cor.Y.glm 1 1.447
## Low.cor.X.glm 1 2.727
## All.X.glm 1 5.848
## All.X.bayesglm 1 4.142
## All.X.no.rnorm.rpart 3 4.588
## All.X.no.rnorm.rf 3 76.855
## SongsLog1.glm 1 3.527
## SongsLog2.glm 1 2.575
## SongsLog3.glm 1 2.638
## min.elapsedtime.final max.auc.fit
## MFO.myMFO_classfr 0.003 0.5000000
## Random.myrandom_classfr 0.002 0.5008601
## Max.cor.Y.cv.0.rpart 0.124 0.5000000
## Max.cor.Y.cv.0.cp.0.rpart 0.113 0.7925561
## Max.cor.Y.rpart 0.126 0.5000000
## Max.cor.Y.glm 0.129 0.6982092
## Interact.High.cor.Y.glm 0.143 0.7238499
## Low.cor.X.glm 0.863 0.8004704
## All.X.glm 0.886 0.8139835
## All.X.bayesglm 1.068 0.8139566
## All.X.no.rnorm.rpart 0.865 0.5000000
## All.X.no.rnorm.rf 25.191 0.9999997
## SongsLog1.glm 0.827 0.8138747
## SongsLog2.glm 0.803 0.8003625
## SongsLog3.glm 0.812 0.8115728
## opt.prob.threshold.fit max.f.score.fit
## MFO.myMFO_classfr 0.5 0.0000000
## Random.myrandom_classfr 0.1 0.2566275
## Max.cor.Y.cv.0.rpart 0.5 0.0000000
## Max.cor.Y.cv.0.cp.0.rpart 0.3 0.4703222
## Max.cor.Y.rpart 0.5 0.0000000
## Max.cor.Y.glm 0.2 0.3323678
## Interact.High.cor.Y.glm 0.2 0.3553903
## Low.cor.X.glm 0.2 0.4514107
## All.X.glm 0.2 0.4666902
## All.X.bayesglm 0.2 0.4650670
## All.X.no.rnorm.rpart 0.5 0.0000000
## All.X.no.rnorm.rf 0.4 0.9990575
## SongsLog1.glm 0.2 0.4704634
## SongsLog2.glm 0.2 0.4508711
## SongsLog3.glm 0.3 0.4644550
## max.Accuracy.fit max.AccuracyLower.fit
## MFO.myMFO_classfr 0.8527982 0.8444027
## Random.myrandom_classfr 0.1472018 0.1390930
## Max.cor.Y.cv.0.rpart 0.8527982 0.8444027
## Max.cor.Y.cv.0.cp.0.rpart 0.8698792 0.8618892
## Max.cor.Y.rpart 0.8328024 0.8444027
## Max.cor.Y.glm 0.8514097 0.7336766
## Interact.High.cor.Y.glm 0.8525202 0.7491514
## Low.cor.X.glm 0.8604359 0.7715504
## All.X.glm 0.8604360 0.7802951
## All.X.bayesglm 0.8605750 0.7798718
## All.X.no.rnorm.rpart 0.8527983 0.8444027
## All.X.no.rnorm.rf 0.8647411 0.9989971
## SongsLog1.glm 0.8609916 0.7825528
## SongsLog2.glm 0.8601582 0.7714094
## SongsLog3.glm 0.8605750 0.8344684
## max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## MFO.myMFO_classfr 0.8609070 0.000000000 0.5000000
## Random.myrandom_classfr 0.1555973 0.000000000 0.5167872
## Max.cor.Y.cv.0.rpart 0.8609070 0.000000000 0.5000000
## Max.cor.Y.cv.0.cp.0.rpart 0.8775683 0.399462097 0.6506801
## Max.cor.Y.rpart 0.8609070 0.040537462 0.5000000
## Max.cor.Y.glm 0.7539745 0.005126623 0.7064126
## Interact.High.cor.Y.glm 0.7690385 0.058007365 0.7395552
## Low.cor.X.glm 0.7907822 0.208371432 0.8415740
## All.X.glm 0.7992502 0.228606299 0.8421138
## All.X.bayesglm 0.7988406 0.225649656 0.8423297
## All.X.no.rnorm.rpart 0.8609070 0.000000000 0.5000000
## All.X.no.rnorm.rf 0.9999664 0.181331612 0.8103206
## SongsLog1.glm 0.8014345 0.228437117 0.8425996
## SongsLog2.glm 0.7906455 0.206782636 0.8430854
## SongsLog3.glm 0.8514076 0.218926492 0.8489690
## opt.prob.threshold.OOB max.f.score.OOB
## MFO.myMFO_classfr 0.5 0.0000000
## Random.myrandom_classfr 0.1 0.2731481
## Max.cor.Y.cv.0.rpart 0.5 0.0000000
## Max.cor.Y.cv.0.cp.0.rpart 0.1 0.3320755
## Max.cor.Y.rpart 0.5 0.0000000
## Max.cor.Y.glm 0.1 0.3312102
## Interact.High.cor.Y.glm 0.2 0.3851852
## Low.cor.X.glm 0.2 0.5064935
## All.X.glm 0.2 0.5222930
## All.X.bayesglm 0.2 0.5222930
## All.X.no.rnorm.rpart 0.5 0.0000000
## All.X.no.rnorm.rf 0.2 0.4767442
## SongsLog1.glm 0.2 0.5161290
## SongsLog2.glm 0.2 0.5161290
## SongsLog3.glm 0.3 0.5087719
## max.Accuracy.OOB max.AccuracyLower.OOB
## MFO.myMFO_classfr 0.8418231 0.8007587
## Random.myrandom_classfr 0.1581769 0.1226449
## Max.cor.Y.cv.0.rpart 0.8418231 0.8007587
## Max.cor.Y.cv.0.cp.0.rpart 0.5254692 0.4734236
## Max.cor.Y.rpart 0.8418231 0.8007587
## Max.cor.Y.glm 0.4369973 0.3859956
## Interact.High.cor.Y.glm 0.7774799 0.7318099
## Low.cor.X.glm 0.7962466 0.7517471
## All.X.glm 0.7989276 0.7546059
## All.X.bayesglm 0.7989276 0.7546059
## All.X.no.rnorm.rpart 0.8418231 0.8007587
## All.X.no.rnorm.rf 0.7587131 0.7119951
## SongsLog1.glm 0.7989276 0.7546059
## SongsLog2.glm 0.7989276 0.7546059
## SongsLog3.glm 0.8498660 0.8095092
## max.AccuracyUpper.OOB max.Kappa.OOB
## MFO.myMFO_classfr 0.8773551 0.00000000
## Random.myrandom_classfr 0.1992413 0.00000000
## Max.cor.Y.cv.0.rpart 0.8773551 0.00000000
## Max.cor.Y.cv.0.cp.0.rpart 0.5771074 0.11425198
## Max.cor.Y.rpart 0.8773551 0.00000000
## Max.cor.Y.glm 0.4890069 0.09998621
## Interact.High.cor.Y.glm 0.8186831 0.25196318
## Low.cor.X.glm 0.8359713 0.38683163
## All.X.glm 0.8384302 0.40474924
## All.X.bayesglm 0.8384302 0.40474924
## All.X.no.rnorm.rpart 0.8773551 0.00000000
## All.X.no.rnorm.rf 0.8012709 0.33945929
## SongsLog1.glm 0.8384302 0.39821886
## SongsLog2.glm 0.8384302 0.39821886
## SongsLog3.glm 0.8845549 0.42029307
## max.AccuracySD.fit max.KappaSD.fit min.aic.fit
## MFO.myMFO_classfr NA NA NA
## Random.myrandom_classfr NA NA NA
## Max.cor.Y.cv.0.rpart NA NA NA
## Max.cor.Y.cv.0.cp.0.rpart NA NA NA
## Max.cor.Y.rpart 0.0097891603 0.020586079 NA
## Max.cor.Y.glm 0.0010199318 0.008109304 5586.714
## Interact.High.cor.Y.glm 0.0014784431 0.017599489 5469.265
## Low.cor.X.glm 0.0008623889 0.005280039 4939.171
## All.X.glm 0.0016670046 0.017700517 4828.367
## All.X.bayesglm 0.0017256879 0.017582275 4828.409
## All.X.no.rnorm.rpart 0.0002050944 0.000000000 NA
## All.X.no.rnorm.rf 0.0010224674 0.012369122 NA
## SongsLog1.glm 0.0021308656 0.015762223 4827.154
## SongsLog2.glm 0.0006681795 0.005733383 4937.849
## SongsLog3.glm 0.0006114180 0.018426137 4848.719rm(ret_lst)
fit.models_1_chunk_df <- myadd_chunk(fit.models_1_chunk_df, "fit.models_1_end",
major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 5 fit.models_1_rf 5 0 126.234 239.861 113.627
## 6 fit.models_1_end 6 0 239.862 NA NAglb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 11 fit.models 7 1 85.439 239.869 154.43
## 12 fit.models 7 2 239.869 NA NAif (!is.null(glb_model_metric_smmry)) {
stats_df <- glb_models_df[, "model_id", FALSE]
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_fitobs_df, glb_rsp_var,
glb_rsp_var_out, model_id, "fit",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_OOBobs_df, glb_rsp_var,
glb_rsp_var_out, model_id, "OOB",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
print("Merging following data into glb_models_df:")
print(stats_mrg_df <- stats_df[, c(1, grep(glb_model_metric, names(stats_df)))])
print(tmp_models_df <- orderBy(~model_id, glb_models_df[, c("model_id",
grep(glb_model_metric, names(stats_df), value=TRUE))]))
tmp2_models_df <- glb_models_df[, c("model_id", setdiff(names(glb_models_df),
grep(glb_model_metric, names(stats_df), value=TRUE)))]
tmp3_models_df <- merge(tmp2_models_df, stats_mrg_df, all.x=TRUE, sort=FALSE)
print(tmp3_models_df)
print(names(tmp3_models_df))
print(glb_models_df <- subset(tmp3_models_df, select=-model_id.1))
}
plt_models_df <- glb_models_df[, -grep("SD|Upper|Lower", names(glb_models_df))]
for (var in grep("^min.", names(plt_models_df), value=TRUE)) {
plt_models_df[, sub("min.", "inv.", var)] <-
#ifelse(all(is.na(tmp <- plt_models_df[, var])), NA, 1.0 / tmp)
1.0 / plt_models_df[, var]
plt_models_df <- plt_models_df[ , -grep(var, names(plt_models_df))]
}
print(plt_models_df)## model_id model_method
## MFO.myMFO_classfr MFO.myMFO_classfr myMFO_classfr
## Random.myrandom_classfr Random.myrandom_classfr myrandom_classfr
## Max.cor.Y.cv.0.rpart Max.cor.Y.cv.0.rpart rpart
## Max.cor.Y.cv.0.cp.0.rpart Max.cor.Y.cv.0.cp.0.rpart rpart
## Max.cor.Y.rpart Max.cor.Y.rpart rpart
## Max.cor.Y.glm Max.cor.Y.glm glm
## Interact.High.cor.Y.glm Interact.High.cor.Y.glm glm
## Low.cor.X.glm Low.cor.X.glm glm
## All.X.glm All.X.glm glm
## All.X.bayesglm All.X.bayesglm bayesglm
## All.X.no.rnorm.rpart All.X.no.rnorm.rpart rpart
## All.X.no.rnorm.rf All.X.no.rnorm.rf rf
## SongsLog1.glm SongsLog1.glm glm
## SongsLog2.glm SongsLog2.glm glm
## SongsLog3.glm SongsLog3.glm glm
## feats
## MFO.myMFO_classfr .rnorm
## Random.myrandom_classfr .rnorm
## Max.cor.Y.cv.0.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.cv.0.cp.0.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.rpart timbre_6_min, timbre_4_max
## Max.cor.Y.glm timbre_6_min, timbre_4_max
## Interact.High.cor.Y.glm timbre_6_min, timbre_4_max, timbre_6_min:timbre_0_max
## Low.cor.X.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.bayesglm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, .rnorm, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.no.rnorm.rpart timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## All.X.no.rnorm.rf timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog1.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog2.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, timbre_9_min, timbre_8_min, energy, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## SongsLog3.glm timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns max.auc.fit
## MFO.myMFO_classfr 0 0.5000000
## Random.myrandom_classfr 0 0.5008601
## Max.cor.Y.cv.0.rpart 0 0.5000000
## Max.cor.Y.cv.0.cp.0.rpart 0 0.7925561
## Max.cor.Y.rpart 3 0.5000000
## Max.cor.Y.glm 1 0.6982092
## Interact.High.cor.Y.glm 1 0.7238499
## Low.cor.X.glm 1 0.8004704
## All.X.glm 1 0.8139835
## All.X.bayesglm 1 0.8139566
## All.X.no.rnorm.rpart 3 0.5000000
## All.X.no.rnorm.rf 3 0.9999997
## SongsLog1.glm 1 0.8138747
## SongsLog2.glm 1 0.8003625
## SongsLog3.glm 1 0.8115728
## opt.prob.threshold.fit max.f.score.fit
## MFO.myMFO_classfr 0.5 0.0000000
## Random.myrandom_classfr 0.1 0.2566275
## Max.cor.Y.cv.0.rpart 0.5 0.0000000
## Max.cor.Y.cv.0.cp.0.rpart 0.3 0.4703222
## Max.cor.Y.rpart 0.5 0.0000000
## Max.cor.Y.glm 0.2 0.3323678
## Interact.High.cor.Y.glm 0.2 0.3553903
## Low.cor.X.glm 0.2 0.4514107
## All.X.glm 0.2 0.4666902
## All.X.bayesglm 0.2 0.4650670
## All.X.no.rnorm.rpart 0.5 0.0000000
## All.X.no.rnorm.rf 0.4 0.9990575
## SongsLog1.glm 0.2 0.4704634
## SongsLog2.glm 0.2 0.4508711
## SongsLog3.glm 0.3 0.4644550
## max.Accuracy.fit max.Kappa.fit max.auc.OOB
## MFO.myMFO_classfr 0.8527982 0.000000000 0.5000000
## Random.myrandom_classfr 0.1472018 0.000000000 0.5167872
## Max.cor.Y.cv.0.rpart 0.8527982 0.000000000 0.5000000
## Max.cor.Y.cv.0.cp.0.rpart 0.8698792 0.399462097 0.6506801
## Max.cor.Y.rpart 0.8328024 0.040537462 0.5000000
## Max.cor.Y.glm 0.8514097 0.005126623 0.7064126
## Interact.High.cor.Y.glm 0.8525202 0.058007365 0.7395552
## Low.cor.X.glm 0.8604359 0.208371432 0.8415740
## All.X.glm 0.8604360 0.228606299 0.8421138
## All.X.bayesglm 0.8605750 0.225649656 0.8423297
## All.X.no.rnorm.rpart 0.8527983 0.000000000 0.5000000
## All.X.no.rnorm.rf 0.8647411 0.181331612 0.8103206
## SongsLog1.glm 0.8609916 0.228437117 0.8425996
## SongsLog2.glm 0.8601582 0.206782636 0.8430854
## SongsLog3.glm 0.8605750 0.218926492 0.8489690
## opt.prob.threshold.OOB max.f.score.OOB
## MFO.myMFO_classfr 0.5 0.0000000
## Random.myrandom_classfr 0.1 0.2731481
## Max.cor.Y.cv.0.rpart 0.5 0.0000000
## Max.cor.Y.cv.0.cp.0.rpart 0.1 0.3320755
## Max.cor.Y.rpart 0.5 0.0000000
## Max.cor.Y.glm 0.1 0.3312102
## Interact.High.cor.Y.glm 0.2 0.3851852
## Low.cor.X.glm 0.2 0.5064935
## All.X.glm 0.2 0.5222930
## All.X.bayesglm 0.2 0.5222930
## All.X.no.rnorm.rpart 0.5 0.0000000
## All.X.no.rnorm.rf 0.2 0.4767442
## SongsLog1.glm 0.2 0.5161290
## SongsLog2.glm 0.2 0.5161290
## SongsLog3.glm 0.3 0.5087719
## max.Accuracy.OOB max.Kappa.OOB
## MFO.myMFO_classfr 0.8418231 0.00000000
## Random.myrandom_classfr 0.1581769 0.00000000
## Max.cor.Y.cv.0.rpart 0.8418231 0.00000000
## Max.cor.Y.cv.0.cp.0.rpart 0.5254692 0.11425198
## Max.cor.Y.rpart 0.8418231 0.00000000
## Max.cor.Y.glm 0.4369973 0.09998621
## Interact.High.cor.Y.glm 0.7774799 0.25196318
## Low.cor.X.glm 0.7962466 0.38683163
## All.X.glm 0.7989276 0.40474924
## All.X.bayesglm 0.7989276 0.40474924
## All.X.no.rnorm.rpart 0.8418231 0.00000000
## All.X.no.rnorm.rf 0.7587131 0.33945929
## SongsLog1.glm 0.7989276 0.39821886
## SongsLog2.glm 0.7989276 0.39821886
## SongsLog3.glm 0.8498660 0.42029307
## inv.elapsedtime.everything inv.elapsedtime.final
## MFO.myMFO_classfr 3.14465409 333.33333333
## Random.myrandom_classfr 4.25531915 500.00000000
## Max.cor.Y.cv.0.rpart 1.18623962 8.06451613
## Max.cor.Y.cv.0.cp.0.rpart 1.78890877 8.84955752
## Max.cor.Y.rpart 0.59347181 7.93650794
## Max.cor.Y.glm 0.75642965 7.75193798
## Interact.High.cor.Y.glm 0.69108500 6.99300699
## Low.cor.X.glm 0.36670334 1.15874855
## All.X.glm 0.17099863 1.12866817
## All.X.bayesglm 0.24142926 0.93632959
## All.X.no.rnorm.rpart 0.21795990 1.15606936
## All.X.no.rnorm.rf 0.01301152 0.03969672
## SongsLog1.glm 0.28352708 1.20918984
## SongsLog2.glm 0.38834951 1.24533001
## SongsLog3.glm 0.37907506 1.23152709
## inv.aic.fit
## MFO.myMFO_classfr NA
## Random.myrandom_classfr NA
## Max.cor.Y.cv.0.rpart NA
## Max.cor.Y.cv.0.cp.0.rpart NA
## Max.cor.Y.rpart NA
## Max.cor.Y.glm 0.0001789961
## Interact.High.cor.Y.glm 0.0001828399
## Low.cor.X.glm 0.0002024631
## All.X.glm 0.0002071093
## All.X.bayesglm 0.0002071076
## All.X.no.rnorm.rpart NA
## All.X.no.rnorm.rf NA
## SongsLog1.glm 0.0002071614
## SongsLog2.glm 0.0002025173
## SongsLog3.glm 0.0002062400print(myplot_radar(radar_inp_df=plt_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 15. Consider specifying shapes manually if you must have them.## Warning: Removed 4 rows containing missing values (geom_path).## Warning: Removed 132 rows containing missing values (geom_point).## Warning: Removed 7 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 15. Consider specifying shapes manually if you must have them.
# print(myplot_radar(radar_inp_df=subset(plt_models_df,
# !(model_id %in% grep("random|MFO", plt_models_df$model_id, value=TRUE)))))
# Compute CI for <metric>SD
glb_models_df <- mutate(glb_models_df,
max.df = ifelse(max.nTuningRuns > 1, max.nTuningRuns - 1, NA),
min.sd2ci.scaler = ifelse(is.na(max.df), NA, qt(0.975, max.df)))
for (var in grep("SD", names(glb_models_df), value=TRUE)) {
# Does CI alredy exist ?
var_components <- unlist(strsplit(var, "SD"))
varActul <- paste0(var_components[1], var_components[2])
varUpper <- paste0(var_components[1], "Upper", var_components[2])
varLower <- paste0(var_components[1], "Lower", var_components[2])
if (varUpper %in% names(glb_models_df)) {
warning(varUpper, " already exists in glb_models_df")
# Assuming Lower also exists
next
}
print(sprintf("var:%s", var))
# CI is dependent on sample size in t distribution; df=n-1
glb_models_df[, varUpper] <- glb_models_df[, varActul] +
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
glb_models_df[, varLower] <- glb_models_df[, varActul] -
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
}## Warning: max.AccuracyUpper.fit already exists in glb_models_df## [1] "var:max.KappaSD.fit"# Plot metrics with CI
plt_models_df <- glb_models_df[, "model_id", FALSE]
pltCI_models_df <- glb_models_df[, "model_id", FALSE]
for (var in grep("Upper", names(glb_models_df), value=TRUE)) {
var_components <- unlist(strsplit(var, "Upper"))
col_name <- unlist(paste(var_components, collapse=""))
plt_models_df[, col_name] <- glb_models_df[, col_name]
for (name in paste0(var_components[1], c("Upper", "Lower"), var_components[2]))
pltCI_models_df[, name] <- glb_models_df[, name]
}
build_statsCI_data <- function(plt_models_df) {
mltd_models_df <- melt(plt_models_df, id.vars="model_id")
mltd_models_df$data <- sapply(1:nrow(mltd_models_df),
function(row_ix) tail(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), "[.]")), 1))
mltd_models_df$label <- sapply(1:nrow(mltd_models_df),
function(row_ix) head(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]),
paste0(".", mltd_models_df[row_ix, "data"]))), 1))
#print(mltd_models_df)
return(mltd_models_df)
}
mltd_models_df <- build_statsCI_data(plt_models_df)
mltdCI_models_df <- melt(pltCI_models_df, id.vars="model_id")
for (row_ix in 1:nrow(mltdCI_models_df)) {
for (type in c("Upper", "Lower")) {
if (length(var_components <- unlist(strsplit(
as.character(mltdCI_models_df[row_ix, "variable"]), type))) > 1) {
#print(sprintf("row_ix:%d; type:%s; ", row_ix, type))
mltdCI_models_df[row_ix, "label"] <- var_components[1]
mltdCI_models_df[row_ix, "data"] <-
unlist(strsplit(var_components[2], "[.]"))[2]
mltdCI_models_df[row_ix, "type"] <- type
break
}
}
}
wideCI_models_df <- reshape(subset(mltdCI_models_df, select=-variable),
timevar="type",
idvar=setdiff(names(mltdCI_models_df), c("type", "value", "variable")),
direction="wide")
#print(wideCI_models_df)
mrgdCI_models_df <- merge(wideCI_models_df, mltd_models_df, all.x=TRUE)
#print(mrgdCI_models_df)
# Merge stats back in if CIs don't exist
goback_vars <- c()
for (var in unique(mltd_models_df$label)) {
for (type in unique(mltd_models_df$data)) {
var_type <- paste0(var, ".", type)
# if this data is already present, next
if (var_type %in% unique(paste(mltd_models_df$label, mltd_models_df$data,
sep=".")))
next
#print(sprintf("var_type:%s", var_type))
goback_vars <- c(goback_vars, var_type)
}
}
if (length(goback_vars) > 0) {
mltd_goback_df <- build_statsCI_data(glb_models_df[, c("model_id", goback_vars)])
mltd_models_df <- rbind(mltd_models_df, mltd_goback_df)
}
mltd_models_df <- merge(mltd_models_df, glb_models_df[, c("model_id", "model_method")],
all.x=TRUE)
png(paste0(glb_out_pfx, "models_bar.png"), width=480*3, height=480*2)
print(gp <- myplot_bar(mltd_models_df, "model_id", "value", colorcol_name="model_method") +
geom_errorbar(data=mrgdCI_models_df,
mapping=aes(x=model_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
facet_grid(label ~ data, scales="free") +
theme(axis.text.x = element_text(angle = 90,vjust = 0.5)))
dev.off()## quartz_off_screen
## 2print(gp)
# used for console inspection
model_evl_terms <- c(NULL)
for (metric in glb_model_evl_criteria)
model_evl_terms <- c(model_evl_terms,
ifelse(length(grep("max", metric)) > 0, "-", "+"), metric)
if (glb_is_classification && glb_is_binomial)
model_evl_terms <- c(model_evl_terms, "-", "opt.prob.threshold.OOB")
model_sel_frmla <- as.formula(paste(c("~ ", model_evl_terms), collapse=" "))
dsp_models_cols <- c("model_id", glb_model_evl_criteria)
if (glb_is_classification && glb_is_binomial)
dsp_models_cols <- c(dsp_models_cols, "opt.prob.threshold.OOB")
print(dsp_models_df <- orderBy(model_sel_frmla, glb_models_df)[, dsp_models_cols])## model_id max.Accuracy.OOB max.auc.OOB max.Kappa.OOB
## 15 SongsLog3.glm 0.8498660 0.8489690 0.42029307
## 1 MFO.myMFO_classfr 0.8418231 0.5000000 0.00000000
## 3 Max.cor.Y.cv.0.rpart 0.8418231 0.5000000 0.00000000
## 5 Max.cor.Y.rpart 0.8418231 0.5000000 0.00000000
## 11 All.X.no.rnorm.rpart 0.8418231 0.5000000 0.00000000
## 14 SongsLog2.glm 0.7989276 0.8430854 0.39821886
## 13 SongsLog1.glm 0.7989276 0.8425996 0.39821886
## 10 All.X.bayesglm 0.7989276 0.8423297 0.40474924
## 9 All.X.glm 0.7989276 0.8421138 0.40474924
## 8 Low.cor.X.glm 0.7962466 0.8415740 0.38683163
## 7 Interact.High.cor.Y.glm 0.7774799 0.7395552 0.25196318
## 12 All.X.no.rnorm.rf 0.7587131 0.8103206 0.33945929
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.5254692 0.6506801 0.11425198
## 6 Max.cor.Y.glm 0.4369973 0.7064126 0.09998621
## 2 Random.myrandom_classfr 0.1581769 0.5167872 0.00000000
## min.aic.fit opt.prob.threshold.OOB
## 15 4848.719 0.3
## 1 NA 0.5
## 3 NA 0.5
## 5 NA 0.5
## 11 NA 0.5
## 14 4937.849 0.2
## 13 4827.154 0.2
## 10 4828.409 0.2
## 9 4828.367 0.2
## 8 4939.171 0.2
## 7 5469.265 0.2
## 12 NA 0.2
## 4 NA 0.1
## 6 5586.714 0.1
## 2 NA 0.1print(myplot_radar(radar_inp_df=dsp_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 15. Consider specifying shapes manually if you must have them.## Warning: Removed 56 rows containing missing values (geom_point).## Warning: Removed 7 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 15. Consider specifying shapes manually if you must have them.
print("Metrics used for model selection:"); print(model_sel_frmla)## [1] "Metrics used for model selection:"## ~-max.Accuracy.OOB - max.auc.OOB - max.Kappa.OOB + min.aic.fit -
## opt.prob.threshold.OOBprint(sprintf("Best model id: %s", dsp_models_df[1, "model_id"]))## [1] "Best model id: SongsLog3.glm"if (is.null(glb_sel_mdl_id)) {
glb_sel_mdl_id <- dsp_models_df[1, "model_id"]
# if (glb_sel_mdl_id == "Interact.High.cor.Y.glm") {
# warning("glb_sel_mdl_id: Interact.High.cor.Y.glm; myextract_mdl_feats does not currently support interaction terms")
# glb_sel_mdl_id <- dsp_models_df[2, "model_id"]
# }
} else
print(sprintf("User specified selection: %s", glb_sel_mdl_id))
myprint_mdl(glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]])
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9182 -0.5417 -0.3481 -0.1874 3.4171
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.196e+01 1.714e+00 6.977 3.01e-12 ***
## timbre_4_max 6.467e-03 1.541e-03 4.196 2.72e-05 ***
## timbre_8_max 4.658e-03 2.988e-03 1.559 0.119022
## timbre_9_max 1.342e-03 2.424e-03 0.554 0.579900
## timbre_11_max 1.984e-02 3.365e-03 5.896 3.74e-09 ***
## timbre_6_max 3.447e-03 2.182e-03 1.580 0.114203
## timbre_1_max -5.115e-04 7.110e-04 -0.719 0.471928
## timbre_10_max 5.793e-03 1.759e-03 3.294 0.000988 ***
## timbre_2_max 4.119e-04 9.020e-04 0.457 0.647915
## tempo_confidence 3.841e-01 1.398e-01 2.747 0.006019 **
## timbre_5_max 2.979e-04 7.855e-04 0.379 0.704526
## timbre_0_min 2.205e-02 4.239e-03 5.200 1.99e-07 ***
## timesignature_confidence 7.143e-01 1.946e-01 3.670 0.000242 ***
## timesignature 1.151e-01 8.726e-02 1.319 0.187183
## timbre_7_max -3.394e-03 1.820e-03 -1.865 0.062208 .
## key 1.649e-02 1.035e-02 1.593 0.111056
## key_confidence 3.394e-01 1.409e-01 2.409 0.015984 *
## timbre_10_min 4.050e-03 1.827e-03 2.217 0.026637 *
## timbre_1_min 5.416e-03 7.643e-04 7.086 1.38e-12 ***
## tempo -6.460e-04 1.665e-03 -0.388 0.698107
## timbre_4_min 1.105e-02 1.978e-03 5.585 2.34e-08 ***
## timbre_3_min 3.179e-04 5.869e-04 0.542 0.588083
## timbre_3_max -2.964e-03 5.758e-04 -5.147 2.64e-07 ***
## loudness 2.306e-01 2.528e-02 9.120 < 2e-16 ***
## timbre_9_min -9.318e-05 2.957e-03 -0.032 0.974859
## timbre_8_min 3.686e-03 2.833e-03 1.301 0.193229
## timbre_5_min -5.135e-03 1.269e-03 -4.046 5.21e-05 ***
## pitch -5.328e+01 6.733e+00 -7.914 2.49e-15 ***
## timbre_2_min -2.254e-03 1.120e-03 -2.012 0.044190 *
## timbre_0_max -3.105e-01 2.537e-02 -12.240 < 2e-16 ***
## timbre_7_min -5.128e-03 1.768e-03 -2.900 0.003733 **
## timbre_11_min -2.638e-02 3.683e-03 -7.162 7.96e-13 ***
## timbre_6_min -1.784e-02 2.246e-03 -7.945 1.94e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4782.7 on 7168 degrees of freedom
## AIC: 4848.7
##
## Number of Fisher Scoring iterations: 6## [1] TRUE# From here to save(), this should all be in one function
# these are executed in the same seq twice more:
# fit.data.training & predict.data.new chunks
glb_get_predictions <- function(df, mdl_id, rsp_var_out, prob_threshold_def=NULL) {
mdl <- glb_models_lst[[mdl_id]]
rsp_var_out <- paste0(rsp_var_out, mdl_id)
if (glb_is_regression) {
df[, rsp_var_out] <- predict(mdl, newdata=df, type="raw")
print(myplot_scatter(df, glb_rsp_var, rsp_var_out, smooth=TRUE))
df[, paste0(rsp_var_out, ".err")] <-
abs(df[, rsp_var_out] - df[, glb_rsp_var])
print(head(orderBy(reformulate(c("-", paste0(rsp_var_out, ".err"))),
df)))
}
if (glb_is_classification && glb_is_binomial) {
prob_threshold <- glb_models_df[glb_models_df$model_id == mdl_id,
"opt.prob.threshold.OOB"]
if (is.null(prob_threshold) || is.na(prob_threshold)) {
warning("Using default probability threshold: ", prob_threshold_def)
if (is.null(prob_threshold <- prob_threshold_def))
stop("Default probability threshold is NULL")
}
df[, paste0(rsp_var_out, ".prob")] <-
predict(mdl, newdata=df, type="prob")[, 2]
df[, rsp_var_out] <-
factor(levels(df[, glb_rsp_var])[
(df[, paste0(rsp_var_out, ".prob")] >=
prob_threshold) * 1 + 1], levels(df[, glb_rsp_var]))
# prediction stats already reported by myfit_mdl ???
}
if (glb_is_classification && !glb_is_binomial) {
df[, rsp_var_out] <- predict(mdl, newdata=df, type="raw")
df[, paste0(rsp_var_out, ".prob")] <-
predict(mdl, newdata=df, type="prob")
}
return(df)
}
glb_OOBobs_df <- glb_get_predictions(df=glb_OOBobs_df, mdl_id=glb_sel_mdl_id,
rsp_var_out=glb_rsp_var_out)
predct_accurate_var_name <- paste0(glb_rsp_var_out, glb_sel_mdl_id, ".accurate")
glb_OOBobs_df[, predct_accurate_var_name] <-
(glb_OOBobs_df[, glb_rsp_var] ==
glb_OOBobs_df[, paste0(glb_rsp_var_out, glb_sel_mdl_id)])
#stop(here"); #sav_models_lst <- glb_models_lst; sav_models_df <- glb_models_df
glb_featsimp_df <-
myget_feats_importance(mdl=glb_sel_mdl, featsimp_df=NULL)
glb_featsimp_df[, paste0(glb_sel_mdl_id, ".importance")] <- glb_featsimp_df$importance
print(glb_featsimp_df)## importance SongsLog3.glm.importance
## timbre_0_max 100.000000 100.000000
## loudness 74.443754 74.443754
## timbre_6_min 64.817463 64.817463
## pitch 64.564454 64.564454
## timbre_11_min 58.403687 58.403687
## timbre_1_min 57.786074 57.786074
## timbre_11_max 48.031092 48.031092
## timbre_4_min 45.486626 45.486626
## timbre_0_min 42.334718 42.334718
## timbre_3_max 41.903716 41.903716
## timbre_4_max 34.107417 34.107417
## timbre_5_min 32.883797 32.883797
## timesignature_confidence 29.803231 29.803231
## timbre_10_max 26.721367 26.721367
## timbre_7_min 23.494570 23.494570
## tempo_confidence 22.240124 22.240124
## key_confidence 19.475923 19.475923
## timbre_10_min 17.899251 17.899251
## timbre_2_min 16.224160 16.224160
## timbre_7_max 15.016247 15.016247
## key 12.793693 12.793693
## timbre_6_max 12.679951 12.679951
## timbre_8_max 12.510488 12.510488
## timesignature 10.545236 10.545236
## timbre_8_min 10.398850 10.398850
## timbre_1_max 5.633911 5.633911
## timbre_9_max 4.275753 4.275753
## timbre_3_min 4.178160 4.178160
## timbre_2_max 3.482307 3.482307
## tempo 2.918906 2.918906
## timbre_5_max 2.847983 2.847983
## timbre_9_min 0.000000 0.000000# Used again in fit.data.training & predict.data.new chunks
glb_analytics_diag_plots <- function(obs_df, mdl_id, prob_threshold=NULL) {
featsimp_df <- glb_featsimp_df
featsimp_df$feat <- gsub("`(.*?)`", "\\1", row.names(featsimp_df))
featsimp_df$feat.interact <- gsub("(.*?):(.*)", "\\2", featsimp_df$feat)
featsimp_df$feat <- gsub("(.*?):(.*)", "\\1", featsimp_df$feat)
featsimp_df$feat.interact <- ifelse(featsimp_df$feat.interact == featsimp_df$feat,
NA, featsimp_df$feat.interact)
featsimp_df$feat <- gsub("(.*?)\\.fctr(.*)", "\\1\\.fctr", featsimp_df$feat)
featsimp_df$feat.interact <- gsub("(.*?)\\.fctr(.*)", "\\1\\.fctr", featsimp_df$feat.interact)
featsimp_df <- orderBy(~ -importance.max, summaryBy(importance ~ feat + feat.interact,
data=featsimp_df, FUN=max))
#rex_str=":(.*)"; txt_vctr=tail(featsimp_df$feat); ret_lst <- regexec(rex_str, txt_vctr); ret_lst <- regmatches(txt_vctr, ret_lst); ret_vctr <- sapply(1:length(ret_lst), function(pos_ix) ifelse(length(ret_lst[[pos_ix]]) > 0, ret_lst[[pos_ix]], "")); print(ret_vctr <- ret_vctr[ret_vctr != ""])
if (nrow(featsimp_df) > 5) {
warning("Limiting important feature scatter plots to 5 out of ", nrow(featsimp_df))
featsimp_df <- head(featsimp_df, 5)
}
# if (!all(is.na(featsimp_df$feat.interact)))
# stop("not implemented yet")
rsp_var_out <- paste0(glb_rsp_var_out, mdl_id)
for (var in featsimp_df$feat) {
plot_df <- melt(obs_df, id.vars=var,
measure.vars=c(glb_rsp_var, rsp_var_out))
# if (var == "<feat_name>") print(myplot_scatter(plot_df, var, "value",
# facet_colcol_name="variable") +
# geom_vline(xintercept=<divider_val>, linetype="dotted")) else
print(myplot_scatter(plot_df, var, "value", colorcol_name="variable",
facet_colcol_name="variable", jitter=TRUE) +
guides(color=FALSE))
}
if (glb_is_regression) {
if (nrow(featsimp_df) == 0)
warning("No important features in glb_fin_mdl") else
print(myplot_prediction_regression(df=obs_df,
feat_x=ifelse(nrow(featsimp_df) > 1, featsimp_df$feat[2],
".rownames"),
feat_y=featsimp_df$feat[1],
rsp_var=glb_rsp_var, rsp_var_out=rsp_var_out,
id_vars=glb_id_var)
# + facet_wrap(reformulate(featsimp_df$feat[2])) # if [1 or 2] is a factor
# + geom_point(aes_string(color="<col_name>.fctr")) # to color the plot
)
}
if (glb_is_classification) {
if (nrow(featsimp_df) == 0)
warning("No features in selected model are statistically important")
else print(myplot_prediction_classification(df=obs_df,
feat_x=ifelse(nrow(featsimp_df) > 1, featsimp_df$feat[2],
".rownames"),
feat_y=featsimp_df$feat[1],
rsp_var=glb_rsp_var,
rsp_var_out=rsp_var_out,
id_vars=glb_id_var,
prob_threshold=prob_threshold)
# + geom_hline(yintercept=<divider_val>, linetype = "dotted")
)
}
}
if (glb_is_classification && glb_is_binomial)
glb_analytics_diag_plots(obs_df=glb_OOBobs_df, mdl_id=glb_sel_mdl_id,
prob_threshold=glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]) else
glb_analytics_diag_plots(obs_df=glb_OOBobs_df, mdl_id=glb_sel_mdl_id) ## Warning in glb_analytics_diag_plots(obs_df = glb_OOBobs_df, mdl_id =
## glb_sel_mdl_id, : Limiting important feature scatter plots to 5 out of 32
## [1] "Min/Max Boundaries: "
## .rownames Top10.fctr Top10.fctr.predict.SongsLog3.glm.prob
## 73051 73051 N 0.11552825
## 73711 73711 Y 0.79838164
## 74931 74931 N 0.02205776
## Top10.fctr.predict.SongsLog3.glm
## 73051 N
## 73711 Y
## 74931 N
## Top10.fctr.predict.SongsLog3.glm.accurate
## 73051 TRUE
## 73711 TRUE
## 74931 TRUE
## Top10.fctr.predict.SongsLog3.glm.error .label
## 73051 0 73051
## 73711 0 73711
## 74931 0 74931
## [1] "Inaccurate: "
## .rownames Top10.fctr Top10.fctr.predict.SongsLog3.glm.prob
## 72271 72271 Y 0.03793471
## 72251 72251 Y 0.05658360
## 73721 73721 Y 0.06631789
## 73701 73701 Y 0.07250412
## 75211 75211 Y 0.09579406
## 75361 75361 Y 0.10175173
## Top10.fctr.predict.SongsLog3.glm
## 72271 N
## 72251 N
## 73721 N
## 73701 N
## 75211 N
## 75361 N
## Top10.fctr.predict.SongsLog3.glm.accurate
## 72271 FALSE
## 72251 FALSE
## 73721 FALSE
## 73701 FALSE
## 75211 FALSE
## 75361 FALSE
## Top10.fctr.predict.SongsLog3.glm.error
## 72271 -0.2620653
## 72251 -0.2434164
## 73721 -0.2336821
## 73701 -0.2274959
## 75211 -0.2042059
## 75361 -0.1982483
## .rownames Top10.fctr Top10.fctr.predict.SongsLog3.glm.prob
## 75361 75361 Y 0.1017517
## 73691 73691 Y 0.1597822
## 74101 74101 Y 0.1927471
## 75331 75331 Y 0.2814858
## 73001 73001 N 0.3216391
## 73031 73031 N 0.3564767
## Top10.fctr.predict.SongsLog3.glm
## 75361 N
## 73691 N
## 74101 N
## 75331 N
## 73001 Y
## 73031 Y
## Top10.fctr.predict.SongsLog3.glm.accurate
## 75361 FALSE
## 73691 FALSE
## 74101 FALSE
## 75331 FALSE
## 73001 FALSE
## 73031 FALSE
## Top10.fctr.predict.SongsLog3.glm.error
## 75361 -0.19824827
## 73691 -0.14021778
## 74101 -0.10725295
## 75331 -0.01851418
## 73001 0.02163909
## 73031 0.05647668
## .rownames Top10.fctr Top10.fctr.predict.SongsLog3.glm.prob
## 75441 75441 N 0.4207847
## 74441 74441 N 0.4573506
## 73891 73891 N 0.4914757
## 75291 75291 N 0.4962977
## 74411 74411 N 0.5298895
## 74171 74171 N 0.5439753
## Top10.fctr.predict.SongsLog3.glm
## 75441 Y
## 74441 Y
## 73891 Y
## 75291 Y
## 74411 Y
## 74171 Y
## Top10.fctr.predict.SongsLog3.glm.accurate
## 75441 FALSE
## 74441 FALSE
## 73891 FALSE
## 75291 FALSE
## 74411 FALSE
## 74171 FALSE
## Top10.fctr.predict.SongsLog3.glm.error
## 75441 0.1207847
## 74441 0.1573506
## 73891 0.1914757
## 75291 0.1962977
## 74411 0.2298895
## 74171 0.2439753
# gather predictions from models better than MFO.*
#mdl_id <- "Conditional.X.rf"
#mdl_id <- "Conditional.X.cp.0.rpart"
#mdl_id <- "Conditional.X.rpart"
# glb_OOBobs_df <- glb_get_predictions(df=glb_OOBobs_df, mdl_id,
# glb_rsp_var_out)
# print(t(confusionMatrix(glb_OOBobs_df[, paste0(glb_rsp_var_out, mdl_id)],
# glb_OOBobs_df[, glb_rsp_var])$table))
# FN_OOB_ids <- c(4721, 4020, 693, 92)
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# grep(glb_rsp_var, names(glb_OOBobs_df), value=TRUE)])
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# glb_feats_df$id[1:5]])
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# glb_txt_vars])
write.csv(glb_OOBobs_df[, c(glb_id_var,
grep(glb_rsp_var, names(glb_OOBobs_df), fixed=TRUE, value=TRUE))],
paste0(gsub(".", "_", paste0(glb_out_pfx, glb_sel_mdl_id), fixed=TRUE),
"_OOBobs.csv"), row.names=FALSE)
# print(glb_allobs_df[glb_allobs_df$UniqueID %in% FN_OOB_ids,
# glb_txt_vars])
# dsp_tbl(Headline.contains="[Ee]bola")
# sum(sel_obs(Headline.contains="[Ee]bola"))
# ftable(xtabs(Popular ~ NewsDesk.fctr, data=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,]))
# xtabs(NewsDesk ~ Popular, #Popular ~ NewsDesk.fctr,
# data=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,],
# exclude=NULL)
# print(mycreate_xtab_df(df=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular", "NewsDesk", "SectionName", "SubsectionName")))
# print(mycreate_tbl_df(df=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular", "NewsDesk", "SectionName", "SubsectionName")))
# print(mycreate_tbl_df(df=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,], c("Popular")))
# print(mycreate_tbl_df(df=glb_allobs_df[sel_obs(Headline.contains="[Ee]bola") ,],
# tbl_col_names=c("Popular", "NewsDesk")))
# write.csv(glb_chunks_df, paste0(glb_out_pfx, tail(glb_chunks_df, 1)$label, "_",
# tail(glb_chunks_df, 1)$step_minor, "_chunks1.csv"),
# row.names=FALSE)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 12 fit.models 7 2 239.869 262.786 22.918
## 13 fit.models 7 3 262.787 NA NAprint(setdiff(names(glb_trnobs_df), names(glb_allobs_df)))## character(0)print(setdiff(names(glb_fitobs_df), names(glb_allobs_df)))## character(0)print(setdiff(names(glb_OOBobs_df), names(glb_allobs_df)))## [1] "Top10.fctr.predict.SongsLog3.glm.prob"
## [2] "Top10.fctr.predict.SongsLog3.glm"
## [3] "Top10.fctr.predict.SongsLog3.glm.accurate"for (col in setdiff(names(glb_OOBobs_df), names(glb_allobs_df)))
# Merge or cbind ?
glb_allobs_df[glb_allobs_df$.lcn == "OOB", col] <- glb_OOBobs_df[, col]
print(setdiff(names(glb_newobs_df), names(glb_allobs_df)))## character(0)if (glb_save_envir)
save(glb_feats_df,
glb_allobs_df, #glb_trnobs_df, glb_fitobs_df, glb_OOBobs_df, glb_newobs_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_sel_mdl, glb_sel_mdl_id,
glb_model_type,
file=paste0(glb_out_pfx, "selmdl_dsk.RData"))
#load(paste0(glb_out_pfx, "selmdl_dsk.RData"))
rm(ret_lst)## Warning in rm(ret_lst): object 'ret_lst' not foundreplay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"model.selected")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 13 fit.models 7 3 262.787 267.96 5.174
## 14 fit.data.training 8 0 267.961 NA NAStep 8.0: fit data training
#load(paste0(glb_inp_pfx, "dsk.RData"))
# To create specific models
# glb_fin_mdl_id <- NULL; glb_fin_mdl <- NULL;
# glb_sel_mdl_id <- "Conditional.X.cp.0.rpart";
# glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]]; print(glb_sel_mdl)
if (!is.null(glb_fin_mdl_id) && (glb_fin_mdl_id %in% names(glb_models_lst))) {
warning("Final model same as user selected model")
glb_fin_mdl <- glb_sel_mdl
} else {
# print(mdl_feats_df <- myextract_mdl_feats(sel_mdl=glb_sel_mdl,
# entity_df=glb_fitobs_df))
if ((model_method <- glb_sel_mdl$method) == "custom")
# get actual method from the model_id
model_method <- tail(unlist(strsplit(glb_sel_mdl_id, "[.]")), 1)
tune_finmdl_df <- NULL
if (nrow(glb_sel_mdl$bestTune) > 0) {
for (param in names(glb_sel_mdl$bestTune)) {
#print(sprintf("param: %s", param))
if (glb_sel_mdl$bestTune[1, param] != "none")
tune_finmdl_df <- rbind(tune_finmdl_df,
data.frame(parameter=param,
min=glb_sel_mdl$bestTune[1, param],
max=glb_sel_mdl$bestTune[1, param],
by=1)) # by val does not matter
}
}
# Sync with parameters in mydsutils.R
require(gdata)
ret_lst <- myfit_mdl(model_id="Final", model_method=model_method,
indep_vars_vctr=trim(unlist(strsplit(glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"feats"], "[,]"))),
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnobs_df, OOB_df=NULL,
n_cv_folds=glb_n_cv_folds, tune_models_df=tune_finmdl_df,
# Automate from here
# Issues if glb_sel_mdl$method == "rf" b/c trainControl is "oob"; not "cv"
model_loss_mtrx=glb_model_metric_terms,
model_summaryFunction=glb_sel_mdl$control$summaryFunction,
model_metric=glb_sel_mdl$metric,
model_metric_maximize=glb_sel_mdl$maximize)
glb_fin_mdl <- glb_models_lst[[length(glb_models_lst)]]
glb_fin_mdl_id <- glb_models_df[length(glb_models_lst), "model_id"]
}## [1] "fitting model: Final.glm"
## [1] " indep_vars: timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min"
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.9182 -0.5417 -0.3481 -0.1874 3.4171
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 1.196e+01 1.714e+00 6.977 3.01e-12 ***
## timbre_4_max 6.467e-03 1.541e-03 4.196 2.72e-05 ***
## timbre_8_max 4.658e-03 2.988e-03 1.559 0.119022
## timbre_9_max 1.342e-03 2.424e-03 0.554 0.579900
## timbre_11_max 1.984e-02 3.365e-03 5.896 3.74e-09 ***
## timbre_6_max 3.447e-03 2.182e-03 1.580 0.114203
## timbre_1_max -5.115e-04 7.110e-04 -0.719 0.471928
## timbre_10_max 5.793e-03 1.759e-03 3.294 0.000988 ***
## timbre_2_max 4.119e-04 9.020e-04 0.457 0.647915
## tempo_confidence 3.841e-01 1.398e-01 2.747 0.006019 **
## timbre_5_max 2.979e-04 7.855e-04 0.379 0.704526
## timbre_0_min 2.205e-02 4.239e-03 5.200 1.99e-07 ***
## timesignature_confidence 7.143e-01 1.946e-01 3.670 0.000242 ***
## timesignature 1.151e-01 8.726e-02 1.319 0.187183
## timbre_7_max -3.394e-03 1.820e-03 -1.865 0.062208 .
## key 1.649e-02 1.035e-02 1.593 0.111056
## key_confidence 3.394e-01 1.409e-01 2.409 0.015984 *
## timbre_10_min 4.050e-03 1.827e-03 2.217 0.026637 *
## timbre_1_min 5.416e-03 7.643e-04 7.086 1.38e-12 ***
## tempo -6.460e-04 1.665e-03 -0.388 0.698107
## timbre_4_min 1.105e-02 1.978e-03 5.585 2.34e-08 ***
## timbre_3_min 3.179e-04 5.869e-04 0.542 0.588083
## timbre_3_max -2.964e-03 5.758e-04 -5.147 2.64e-07 ***
## loudness 2.306e-01 2.528e-02 9.120 < 2e-16 ***
## timbre_9_min -9.318e-05 2.957e-03 -0.032 0.974859
## timbre_8_min 3.686e-03 2.833e-03 1.301 0.193229
## timbre_5_min -5.135e-03 1.269e-03 -4.046 5.21e-05 ***
## pitch -5.328e+01 6.733e+00 -7.914 2.49e-15 ***
## timbre_2_min -2.254e-03 1.120e-03 -2.012 0.044190 *
## timbre_0_max -3.105e-01 2.537e-02 -12.240 < 2e-16 ***
## timbre_7_min -5.128e-03 1.768e-03 -2.900 0.003733 **
## timbre_11_min -2.638e-02 3.683e-03 -7.162 7.96e-13 ***
## timbre_6_min -1.784e-02 2.246e-03 -7.945 1.94e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 6017.5 on 7200 degrees of freedom
## Residual deviance: 4782.7 on 7168 degrees of freedom
## AIC: 4848.7
##
## Number of Fisher Scoring iterations: 6
##
## [1] " calling mypredict_mdl for fit:"
## threshold f.score
## 1 0.0 0.25662753
## 2 0.1 0.40791269
## 3 0.2 0.45992300
## 4 0.3 0.46445498
## 5 0.4 0.38008500
## 6 0.5 0.28034682
## 7 0.6 0.18800648
## 8 0.7 0.12000000
## 9 0.8 0.04044118
## 10 0.9 0.01312090
## 11 1.0 0.00000000## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## Top10.fctr Top10.fctr.predict.Final.glm.N Top10.fctr.predict.Final.glm.Y
## 1 N 5581 560
## 2 Y 570 490
## Prediction
## Reference N Y
## N 5581 560
## Y 570 490
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 0.8430774 0.3725277 0.8344684 0.8514076 0.8527982
## AccuracyPValue McnemarPValue
## 0.9900387 0.7889042## Warning in mypredict_mdl(mdl, df = fit_df, rsp_var, rsp_var_out,
## model_id_method, : Expecting 1 metric: Accuracy; recd: Accuracy, Kappa;
## retaining Accuracy only
## model_id model_method
## 1 Final.glm glm
## feats
## 1 timbre_4_max, timbre_8_max, timbre_9_max, timbre_11_max, timbre_6_max, timbre_1_max, timbre_10_max, timbre_2_max, tempo_confidence, timbre_5_max, timbre_0_min, timesignature_confidence, timesignature, timbre_7_max, key, key_confidence, timbre_10_min, timbre_1_min, tempo, timbre_4_min, timbre_3_min, timbre_3_max, loudness, timbre_9_min, timbre_8_min, timbre_5_min, pitch, timbre_2_min, timbre_0_max, timbre_7_min, timbre_11_min, timbre_6_min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.575 0.815
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8115728 0.3 0.464455 0.860575
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit min.aic.fit
## 1 0.8344684 0.8514076 0.2189265 4848.719
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.000611418 0.01842614rm(ret_lst)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc=FALSE)## label step_major step_minor bgn end elapsed
## 14 fit.data.training 8 0 267.961 277.061 9.1
## 15 fit.data.training 8 1 277.062 NA NAglb_trnobs_df <- glb_get_predictions(df=glb_trnobs_df, mdl_id=glb_fin_mdl_id,
rsp_var_out=glb_rsp_var_out,
prob_threshold_def=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id, "opt.prob.threshold.OOB"], NULL))## Warning in glb_get_predictions(df = glb_trnobs_df, mdl_id =
## glb_fin_mdl_id, : Using default probability threshold: 0.3sav_featsimp_df <- glb_featsimp_df
#glb_feats_df <- sav_feats_df
# glb_feats_df <- mymerge_feats_importance(feats_df=glb_feats_df, sel_mdl=glb_fin_mdl,
# entity_df=glb_trnobs_df)
glb_featsimp_df <- myget_feats_importance(mdl=glb_fin_mdl, featsimp_df=glb_featsimp_df)
glb_featsimp_df[, paste0(glb_fin_mdl_id, ".importance")] <- glb_featsimp_df$importance
print(glb_featsimp_df)## SongsLog3.glm.importance importance
## timbre_0_max 100.000000 100.000000
## loudness 74.443754 74.443754
## timbre_6_min 64.817463 64.817463
## pitch 64.564454 64.564454
## timbre_11_min 58.403687 58.403687
## timbre_1_min 57.786074 57.786074
## timbre_11_max 48.031092 48.031092
## timbre_4_min 45.486626 45.486626
## timbre_0_min 42.334718 42.334718
## timbre_3_max 41.903716 41.903716
## timbre_4_max 34.107417 34.107417
## timbre_5_min 32.883797 32.883797
## timesignature_confidence 29.803231 29.803231
## timbre_10_max 26.721367 26.721367
## timbre_7_min 23.494570 23.494570
## tempo_confidence 22.240124 22.240124
## key_confidence 19.475923 19.475923
## timbre_10_min 17.899251 17.899251
## timbre_2_min 16.224160 16.224160
## timbre_7_max 15.016247 15.016247
## key 12.793693 12.793693
## timbre_6_max 12.679951 12.679951
## timbre_8_max 12.510488 12.510488
## timesignature 10.545236 10.545236
## timbre_8_min 10.398850 10.398850
## timbre_1_max 5.633911 5.633911
## timbre_9_max 4.275753 4.275753
## timbre_3_min 4.178160 4.178160
## timbre_2_max 3.482307 3.482307
## tempo 2.918906 2.918906
## timbre_5_max 2.847983 2.847983
## timbre_9_min 0.000000 0.000000
## Final.glm.importance
## timbre_0_max 100.000000
## loudness 74.443754
## timbre_6_min 64.817463
## pitch 64.564454
## timbre_11_min 58.403687
## timbre_1_min 57.786074
## timbre_11_max 48.031092
## timbre_4_min 45.486626
## timbre_0_min 42.334718
## timbre_3_max 41.903716
## timbre_4_max 34.107417
## timbre_5_min 32.883797
## timesignature_confidence 29.803231
## timbre_10_max 26.721367
## timbre_7_min 23.494570
## tempo_confidence 22.240124
## key_confidence 19.475923
## timbre_10_min 17.899251
## timbre_2_min 16.224160
## timbre_7_max 15.016247
## key 12.793693
## timbre_6_max 12.679951
## timbre_8_max 12.510488
## timesignature 10.545236
## timbre_8_min 10.398850
## timbre_1_max 5.633911
## timbre_9_max 4.275753
## timbre_3_min 4.178160
## timbre_2_max 3.482307
## tempo 2.918906
## timbre_5_max 2.847983
## timbre_9_min 0.000000if (glb_is_classification && glb_is_binomial)
glb_analytics_diag_plots(obs_df=glb_trnobs_df, mdl_id=glb_fin_mdl_id,
prob_threshold=glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]) else
glb_analytics_diag_plots(obs_df=glb_trnobs_df, mdl_id=glb_fin_mdl_id) ## Warning in glb_analytics_diag_plots(obs_df = glb_trnobs_df, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 32
## [1] "Min/Max Boundaries: "
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 2147 2147 N 6.851974e-12
## 2307 2307 N 1.215963e-02
## 559 559 N 3.593032e-02
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 2147 N TRUE
## 2307 N TRUE
## 559 N TRUE
## Top10.fctr.predict.Final.glm.error .label
## 2147 0 2147
## 2307 0 2307
## 559 0 559
## [1] "Inaccurate: "
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 5071 5071 Y 0.00291371
## 7485 7485 Y 0.01365728
## 1124 1124 Y 0.01445043
## 6869 6869 Y 0.01842675
## 4901 4901 Y 0.01871727
## 2214 2214 Y 0.01901933
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 5071 N FALSE
## 7485 N FALSE
## 1124 N FALSE
## 6869 N FALSE
## 4901 N FALSE
## 2214 N FALSE
## Top10.fctr.predict.Final.glm.error
## 5071 -0.2970863
## 7485 -0.2863427
## 1124 -0.2855496
## 6869 -0.2815732
## 4901 -0.2812827
## 2214 -0.2809807
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 909 909 Y 0.08826412
## 6194 6194 Y 0.08860972
## 1825 1825 Y 0.15018048
## 1359 1359 N 0.31637964
## 4752 4752 N 0.36445023
## 1471 1471 N 0.36488861
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 909 N FALSE
## 6194 N FALSE
## 1825 N FALSE
## 1359 Y FALSE
## 4752 Y FALSE
## 1471 Y FALSE
## Top10.fctr.predict.Final.glm.error
## 909 -0.21173588
## 6194 -0.21139028
## 1825 -0.14981952
## 1359 0.01637964
## 4752 0.06445023
## 1471 0.06488861
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 5253 5253 N 0.8057353
## 6974 6974 N 0.8071618
## 1278 1278 N 0.8134451
## 3940 3940 N 0.8213381
## 6206 6206 N 0.8394882
## 3680 3680 N 0.8411511
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 5253 Y FALSE
## 6974 Y FALSE
## 1278 Y FALSE
## 3940 Y FALSE
## 6206 Y FALSE
## 3680 Y FALSE
## Top10.fctr.predict.Final.glm.error
## 5253 0.5057353
## 6974 0.5071618
## 1278 0.5134451
## 3940 0.5213381
## 6206 0.5394882
## 3680 0.5411511
dsp_feats_vctr <- c(NULL)
for(var in grep(".importance", names(glb_feats_df), fixed=TRUE, value=TRUE))
dsp_feats_vctr <- union(dsp_feats_vctr,
glb_feats_df[!is.na(glb_feats_df[, var]), "id"])
# print(glb_trnobs_df[glb_trnobs_df$UniqueID %in% FN_OOB_ids,
# grep(glb_rsp_var, names(glb_trnobs_df), value=TRUE)])
print(setdiff(names(glb_trnobs_df), names(glb_allobs_df)))## [1] "Top10.fctr.predict.Final.glm.prob" "Top10.fctr.predict.Final.glm"for (col in setdiff(names(glb_trnobs_df), names(glb_allobs_df)))
# Merge or cbind ?
glb_allobs_df[glb_allobs_df$.src == "Train", col] <- glb_trnobs_df[, col]
print(setdiff(names(glb_fitobs_df), names(glb_allobs_df)))## character(0)print(setdiff(names(glb_OOBobs_df), names(glb_allobs_df)))## character(0)for (col in setdiff(names(glb_OOBobs_df), names(glb_allobs_df)))
# Merge or cbind ?
glb_allobs_df[glb_allobs_df$.lcn == "OOB", col] <- glb_OOBobs_df[, col]
print(setdiff(names(glb_newobs_df), names(glb_allobs_df)))## character(0)if (glb_save_envir)
save(glb_feats_df, glb_allobs_df,
#glb_trnobs_df, glb_fitobs_df, glb_OOBobs_df, glb_newobs_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_model_type,
glb_sel_mdl, glb_sel_mdl_id,
glb_fin_mdl, glb_fin_mdl_id,
file=paste0(glb_out_pfx, "dsk.RData"))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all.prediction","model.final")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
glb_chunks_df <- myadd_chunk(glb_chunks_df, "predict.data.new", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 15 fit.data.training 8 1 277.062 288 10.938
## 16 predict.data.new 9 0 288.000 NA NAStep 9.0: predict data new
# Compute final model predictions
# sav_newobs_df <- glb_newobs_df
glb_newobs_df <- glb_get_predictions(glb_newobs_df, mdl_id=glb_fin_mdl_id,
rsp_var_out=glb_rsp_var_out,
prob_threshold_def=ifelse(glb_is_classification && glb_is_binomial,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"], NULL))## Warning in glb_get_predictions(glb_newobs_df, mdl_id = glb_fin_mdl_id,
## rsp_var_out = glb_rsp_var_out, : Using default probability threshold: 0.3if (glb_is_classification && glb_is_binomial)
glb_analytics_diag_plots(obs_df=glb_newobs_df, mdl_id=glb_fin_mdl_id,
prob_threshold=glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]) else
glb_analytics_diag_plots(obs_df=glb_newobs_df, mdl_id=glb_fin_mdl_id) ## Warning in glb_analytics_diag_plots(obs_df = glb_newobs_df, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 32
## [1] "Min/Max Boundaries: "
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 73051 73051 N 0.11552825
## 73711 73711 Y 0.79838164
## 74931 74931 N 0.02205776
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 73051 N TRUE
## 73711 Y TRUE
## 74931 N TRUE
## Top10.fctr.predict.Final.glm.error .label
## 73051 0 73051
## 73711 0 73711
## 74931 0 74931
## [1] "Inaccurate: "
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 72271 72271 Y 0.03793471
## 72251 72251 Y 0.05658360
## 73721 73721 Y 0.06631789
## 73701 73701 Y 0.07250412
## 75211 75211 Y 0.09579406
## 75361 75361 Y 0.10175173
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 72271 N FALSE
## 72251 N FALSE
## 73721 N FALSE
## 73701 N FALSE
## 75211 N FALSE
## 75361 N FALSE
## Top10.fctr.predict.Final.glm.error
## 72271 -0.2620653
## 72251 -0.2434164
## 73721 -0.2336821
## 73701 -0.2274959
## 75211 -0.2042059
## 75361 -0.1982483
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 72271 72271 Y 0.03793471
## 73101 73101 Y 0.13016697
## 75221 75221 Y 0.22328723
## 73931 73931 N 0.31237727
## 72591 72591 N 0.32940462
## 75291 75291 N 0.49629767
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 72271 N FALSE
## 73101 N FALSE
## 75221 N FALSE
## 73931 Y FALSE
## 72591 Y FALSE
## 75291 Y FALSE
## Top10.fctr.predict.Final.glm.error
## 72271 -0.26206529
## 73101 -0.16983303
## 75221 -0.07671277
## 73931 0.01237727
## 72591 0.02940462
## 75291 0.19629767
## .rownames Top10.fctr Top10.fctr.predict.Final.glm.prob
## 75441 75441 N 0.4207847
## 74441 74441 N 0.4573506
## 73891 73891 N 0.4914757
## 75291 75291 N 0.4962977
## 74411 74411 N 0.5298895
## 74171 74171 N 0.5439753
## Top10.fctr.predict.Final.glm Top10.fctr.predict.Final.glm.accurate
## 75441 Y FALSE
## 74441 Y FALSE
## 73891 Y FALSE
## 75291 Y FALSE
## 74411 Y FALSE
## 74171 Y FALSE
## Top10.fctr.predict.Final.glm.error
## 75441 0.1207847
## 74441 0.1573506
## 73891 0.1914757
## 75291 0.1962977
## 74411 0.2298895
## 74171 0.2439753
if (glb_is_classification && glb_is_binomial) {
submit_df <- glb_newobs_df[, c(glb_id_var,
paste0(glb_rsp_var_out, glb_fin_mdl_id, ".prob"))]
names(submit_df)[2] <- "Probability1"
# submit_df <- glb_newobs_df[, c(paste0(glb_rsp_var_out, glb_fin_mdl_id)), FALSE]
# names(submit_df)[1] <- "BDscience"
# submit_df$BDscience <- as.numeric(submit_df$BDscience) - 1
# #submit_df <-rbind(submit_df, data.frame(bdanalytics=c(" ")))
# print("Submission Stats:")
# print(table(submit_df$BDscience, useNA = "ifany"))
} else submit_df <- glb_newobs_df[, c(glb_id_var,
paste0(glb_rsp_var_out, glb_fin_mdl_id))]
submit_fname <- paste0(gsub(".", "_", paste0(glb_out_pfx, glb_fin_mdl_id), fixed=TRUE),
"_submit.csv")
write.csv(submit_df, submit_fname, quote=FALSE, row.names=FALSE)
#cat(" ", "\n", file=submit_fn, append=TRUE)
# print(orderBy(~ -max.auc.OOB, glb_models_df[, c("model_id",
# "max.auc.OOB", "max.Accuracy.OOB")]))
if (glb_is_classification && glb_is_binomial)
print(glb_models_df[glb_models_df$model_id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"])## [1] 0.3print(sprintf("glb_sel_mdl_id: %s", glb_sel_mdl_id))## [1] "glb_sel_mdl_id: SongsLog3.glm"print(sprintf("glb_fin_mdl_id: %s", glb_fin_mdl_id))## [1] "glb_fin_mdl_id: Final.glm"print(dim(glb_fitobs_df))## [1] 7201 43print(dsp_models_df)## model_id max.Accuracy.OOB max.auc.OOB max.Kappa.OOB
## 15 SongsLog3.glm 0.8498660 0.8489690 0.42029307
## 1 MFO.myMFO_classfr 0.8418231 0.5000000 0.00000000
## 3 Max.cor.Y.cv.0.rpart 0.8418231 0.5000000 0.00000000
## 5 Max.cor.Y.rpart 0.8418231 0.5000000 0.00000000
## 11 All.X.no.rnorm.rpart 0.8418231 0.5000000 0.00000000
## 14 SongsLog2.glm 0.7989276 0.8430854 0.39821886
## 13 SongsLog1.glm 0.7989276 0.8425996 0.39821886
## 10 All.X.bayesglm 0.7989276 0.8423297 0.40474924
## 9 All.X.glm 0.7989276 0.8421138 0.40474924
## 8 Low.cor.X.glm 0.7962466 0.8415740 0.38683163
## 7 Interact.High.cor.Y.glm 0.7774799 0.7395552 0.25196318
## 12 All.X.no.rnorm.rf 0.7587131 0.8103206 0.33945929
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.5254692 0.6506801 0.11425198
## 6 Max.cor.Y.glm 0.4369973 0.7064126 0.09998621
## 2 Random.myrandom_classfr 0.1581769 0.5167872 0.00000000
## min.aic.fit opt.prob.threshold.OOB
## 15 4848.719 0.3
## 1 NA 0.5
## 3 NA 0.5
## 5 NA 0.5
## 11 NA 0.5
## 14 4937.849 0.2
## 13 4827.154 0.2
## 10 4828.409 0.2
## 9 4828.367 0.2
## 8 4939.171 0.2
## 7 5469.265 0.2
## 12 NA 0.2
## 4 NA 0.1
## 6 5586.714 0.1
## 2 NA 0.1if (glb_is_regression) {
print(sprintf("%s OOB RMSE: %0.4f", glb_sel_mdl_id,
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id, "min.RMSE.OOB"]))
if (!is.null(glb_category_vars)) {
stop("not implemented yet")
tmp_OOBobs_df <- glb_OOBobs_df[, c(glb_category_vars, predct_accurate_var_name)]
names(tmp_OOBobs_df)[length(names(tmp_OOBobs_df))] <- "accurate.OOB"
aOOB_ctgry_df <- mycreate_xtab_df(tmp_OOBobs_df, names(tmp_OOBobs_df))
aOOB_ctgry_df[is.na(aOOB_ctgry_df)] <- 0
aOOB_ctgry_df <- mutate(aOOB_ctgry_df,
.n.OOB = accurate.OOB.FALSE + accurate.OOB.TRUE,
max.accuracy.OOB = accurate.OOB.TRUE / .n.OOB)
#intersect(names(glb_ctgry_df), names(aOOB_ctgry_df))
glb_ctgry_df <- merge(glb_ctgry_df, aOOB_ctgry_df, all=TRUE)
print(orderBy(~-accurate.OOB.FALSE, glb_ctgry_df))
}
if ((glb_rsp_var %in% names(glb_newobs_df)) &&
!(any(is.na(glb_newobs_df[, glb_rsp_var])))) {
pred_stats_df <-
mypredict_mdl(mdl=glb_models_lst[[glb_fin_mdl_id]],
df=glb_newobs_df,
rsp_var=glb_rsp_var,
rsp_var_out=glb_rsp_var_out,
model_id_method=glb_fin_mdl_id,
label="new",
model_summaryFunction=glb_sel_mdl$control$summaryFunction,
model_metric=glb_sel_mdl$metric,
model_metric_maximize=glb_sel_mdl$maximize,
ret_type="stats")
print(sprintf("%s prediction stats for glb_newobs_df:", glb_fin_mdl_id))
print(pred_stats_df)
}
}
if (glb_is_classification) {
print(sprintf("%s OOB confusion matrix & accuracy: ", glb_sel_mdl_id))
print(t(confusionMatrix(glb_OOBobs_df[, paste0(glb_rsp_var_out, glb_sel_mdl_id)],
glb_OOBobs_df[, glb_rsp_var])$table))
if (!is.null(glb_category_vars)) {
tmp_OOBobs_df <- glb_OOBobs_df[, c(glb_category_vars, predct_accurate_var_name)]
names(tmp_OOBobs_df)[length(names(tmp_OOBobs_df))] <- "accurate.OOB"
aOOB_ctgry_df <- mycreate_xtab_df(tmp_OOBobs_df, names(tmp_OOBobs_df))
aOOB_ctgry_df[is.na(aOOB_ctgry_df)] <- 0
aOOB_ctgry_df <- mutate(aOOB_ctgry_df,
.n.OOB = accurate.OOB.FALSE + accurate.OOB.TRUE,
max.accuracy.OOB = accurate.OOB.TRUE / .n.OOB)
#intersect(names(glb_ctgry_df), names(aOOB_ctgry_df))
glb_ctgry_df <- merge(glb_ctgry_df, aOOB_ctgry_df, all=TRUE)
print(orderBy(~-accurate.OOB.FALSE, glb_ctgry_df))
}
if ((glb_rsp_var %in% names(glb_newobs_df)) &&
!(any(is.na(glb_newobs_df[, glb_rsp_var])))) {
print(sprintf("%s new confusion matrix & accuracy: ", glb_fin_mdl_id))
print(t(confusionMatrix(glb_newobs_df[, paste0(glb_rsp_var_out, glb_fin_mdl_id)],
glb_newobs_df[, glb_rsp_var])$table))
}
} ## [1] "SongsLog3.glm OOB confusion matrix & accuracy: "
## Prediction
## Reference N Y
## N 288 26
## Y 30 29
## [1] "Final.glm new confusion matrix & accuracy: "
## Prediction
## Reference N Y
## N 288 26
## Y 30 29dsp_myCategory_conf_mtrx <- function(myCategory) {
print(sprintf("%s OOB::myCategory=%s confusion matrix & accuracy: ",
glb_sel_mdl_id, myCategory))
print(t(confusionMatrix(
glb_OOBobs_df[glb_OOBobs_df$myCategory == myCategory,
paste0(glb_rsp_var_out, glb_sel_mdl_id)],
glb_OOBobs_df[glb_OOBobs_df$myCategory == myCategory, glb_rsp_var])$table))
print(sum(glb_OOBobs_df[glb_OOBobs_df$myCategory == myCategory,
predct_accurate_var_name]) /
nrow(glb_OOBobs_df[glb_OOBobs_df$myCategory == myCategory, ]))
err_ids <- glb_OOBobs_df[(glb_OOBobs_df$myCategory == myCategory) &
(!glb_OOBobs_df[, predct_accurate_var_name]), glb_id_var]
OOB_FNerr_df <- glb_OOBobs_df[(glb_OOBobs_df$UniqueID %in% err_ids) &
(glb_OOBobs_df$Popular == 1),
c(
".clusterid",
"Popular", "Headline", "Snippet", "Abstract")]
print(sprintf("%s OOB::myCategory=%s FN errors: %d", glb_sel_mdl_id, myCategory,
nrow(OOB_FNerr_df)))
print(OOB_FNerr_df)
OOB_FPerr_df <- glb_OOBobs_df[(glb_OOBobs_df$UniqueID %in% err_ids) &
(glb_OOBobs_df$Popular == 0),
c(
".clusterid",
"Popular", "Headline", "Snippet", "Abstract")]
print(sprintf("%s OOB::myCategory=%s FP errors: %d", glb_sel_mdl_id, myCategory,
nrow(OOB_FPerr_df)))
print(OOB_FPerr_df)
}
#dsp_myCategory_conf_mtrx(myCategory="OpEd#Opinion#")
#dsp_myCategory_conf_mtrx(myCategory="Business#Business Day#Dealbook")
#dsp_myCategory_conf_mtrx(myCategory="##")
# if (glb_is_classification) {
# print("FN_OOB_ids:")
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# grep(glb_rsp_var, names(glb_OOBobs_df), value=TRUE)])
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# glb_txt_vars])
# print(dsp_vctr <- colSums(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# setdiff(grep("[HSA].", names(glb_OOBobs_df), value=TRUE),
# union(myfind_chr_cols_df(glb_OOBobs_df),
# grep(".fctr", names(glb_OOBobs_df), fixed=TRUE, value=TRUE)))]))
# }
dsp_hdlpfx_results <- function(hdlpfx) {
print(hdlpfx)
print(glb_OOBobs_df[glb_OOBobs_df$Headline.pfx %in% c(hdlpfx),
grep(glb_rsp_var, names(glb_OOBobs_df), value=TRUE)])
print(glb_newobs_df[glb_newobs_df$Headline.pfx %in% c(hdlpfx),
grep(glb_rsp_var, names(glb_newobs_df), value=TRUE)])
print(dsp_vctr <- colSums(glb_newobs_df[glb_newobs_df$Headline.pfx %in% c(hdlpfx),
setdiff(grep("[HSA]\\.", names(glb_newobs_df), value=TRUE),
union(myfind_chr_cols_df(glb_newobs_df),
grep(".fctr", names(glb_newobs_df), fixed=TRUE, value=TRUE)))]))
print(dsp_vctr <- dsp_vctr[dsp_vctr != 0])
print(glb_newobs_df[glb_newobs_df$Headline.pfx %in% c(hdlpfx),
union(names(dsp_vctr), myfind_chr_cols_df(glb_newobs_df))])
}
#dsp_hdlpfx_results(hdlpfx="Ask Well::")
# print("myMisc::|OpEd|blank|blank|1:")
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% c(6446),
# grep(glb_rsp_var, names(glb_OOBobs_df), value=TRUE)])
# print(glb_OOBobs_df[glb_OOBobs_df$UniqueID %in% FN_OOB_ids,
# c("WordCount", "WordCount.log", "myMultimedia",
# "NewsDesk", "SectionName", "SubsectionName")])
# print(mycreate_sqlxtab_df(glb_allobs_df[sel_obs(Headline.contains="[Vv]ideo"), ],
# c(glb_rsp_var, "myMultimedia")))
# dsp_chisq.test(Headline.contains="[Vi]deo")
# print(glb_allobs_df[sel_obs(Headline.contains="[Vv]ideo"),
# c(glb_rsp_var, "Popular", "myMultimedia", "Headline")])
# print(glb_allobs_df[sel_obs(Headline.contains="[Ee]bola", Popular=1),
# c(glb_rsp_var, "Popular", "myMultimedia", "Headline",
# "NewsDesk", "SectionName", "SubsectionName")])
# print(subset(glb_feats_df, !is.na(importance))[,
# c("is.ConditionalX.y",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
# print(subset(glb_feats_df, is.ConditionalX.y & is.na(importance))[,
# c("is.ConditionalX.y",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
# print(subset(glb_feats_df, !is.na(importance))[,
# c("zeroVar", "nzv", "myNearZV",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
# print(subset(glb_feats_df, is.na(importance))[,
# c("zeroVar", "nzv", "myNearZV",
# grep("importance", names(glb_feats_df), fixed=TRUE, value=TRUE))])
print(orderBy(as.formula(paste0("~ -", glb_sel_mdl_id, ".importance")), glb_featsimp_df))## SongsLog3.glm.importance importance
## timbre_0_max 100.000000 100.000000
## loudness 74.443754 74.443754
## timbre_6_min 64.817463 64.817463
## pitch 64.564454 64.564454
## timbre_11_min 58.403687 58.403687
## timbre_1_min 57.786074 57.786074
## timbre_11_max 48.031092 48.031092
## timbre_4_min 45.486626 45.486626
## timbre_0_min 42.334718 42.334718
## timbre_3_max 41.903716 41.903716
## timbre_4_max 34.107417 34.107417
## timbre_5_min 32.883797 32.883797
## timesignature_confidence 29.803231 29.803231
## timbre_10_max 26.721367 26.721367
## timbre_7_min 23.494570 23.494570
## tempo_confidence 22.240124 22.240124
## key_confidence 19.475923 19.475923
## timbre_10_min 17.899251 17.899251
## timbre_2_min 16.224160 16.224160
## timbre_7_max 15.016247 15.016247
## key 12.793693 12.793693
## timbre_6_max 12.679951 12.679951
## timbre_8_max 12.510488 12.510488
## timesignature 10.545236 10.545236
## timbre_8_min 10.398850 10.398850
## timbre_1_max 5.633911 5.633911
## timbre_9_max 4.275753 4.275753
## timbre_3_min 4.178160 4.178160
## timbre_2_max 3.482307 3.482307
## tempo 2.918906 2.918906
## timbre_5_max 2.847983 2.847983
## timbre_9_min 0.000000 0.000000
## Final.glm.importance
## timbre_0_max 100.000000
## loudness 74.443754
## timbre_6_min 64.817463
## pitch 64.564454
## timbre_11_min 58.403687
## timbre_1_min 57.786074
## timbre_11_max 48.031092
## timbre_4_min 45.486626
## timbre_0_min 42.334718
## timbre_3_max 41.903716
## timbre_4_max 34.107417
## timbre_5_min 32.883797
## timesignature_confidence 29.803231
## timbre_10_max 26.721367
## timbre_7_min 23.494570
## tempo_confidence 22.240124
## key_confidence 19.475923
## timbre_10_min 17.899251
## timbre_2_min 16.224160
## timbre_7_max 15.016247
## key 12.793693
## timbre_6_max 12.679951
## timbre_8_max 12.510488
## timesignature 10.545236
## timbre_8_min 10.398850
## timbre_1_max 5.633911
## timbre_9_max 4.275753
## timbre_3_min 4.178160
## timbre_2_max 3.482307
## tempo 2.918906
## timbre_5_max 2.847983
## timbre_9_min 0.000000# players_df <- data.frame(id=c("Chavez", "Giambi", "Menechino", "Myers", "Pena"),
# OBP=c(0.338, 0.391, 0.369, 0.313, 0.361),
# SLG=c(0.540, 0.450, 0.374, 0.447, 0.500),
# cost=c(1400000, 1065000, 295000, 800000, 300000))
# players_df$RS.predict <- predict(glb_models_lst[[csm_mdl_id]], players_df)
# print(orderBy(~ -RS.predict, players_df))
if (length(diff <- setdiff(names(glb_trnobs_df), names(glb_allobs_df))) > 0)
print(diff)
for (col in setdiff(names(glb_trnobs_df), names(glb_allobs_df)))
# Merge or cbind ?
glb_allobs_df[glb_allobs_df$.src == "Train", col] <- glb_trnobs_df[, col]
if (length(diff <- setdiff(names(glb_fitobs_df), names(glb_allobs_df))) > 0)
print(diff)
if (length(diff <- setdiff(names(glb_OOBobs_df), names(glb_allobs_df))) > 0)
print(diff)
for (col in setdiff(names(glb_OOBobs_df), names(glb_allobs_df)))
# Merge or cbind ?
glb_allobs_df[glb_allobs_df$.lcn == "OOB", col] <- glb_OOBobs_df[, col]
if (length(diff <- setdiff(names(glb_newobs_df), names(glb_allobs_df))) > 0)
print(diff)
if (glb_save_envir)
save(glb_feats_df, glb_allobs_df,
#glb_trnobs_df, glb_fitobs_df, glb_OOBobs_df, glb_newobs_df,
glb_models_df, dsp_models_df, glb_models_lst, glb_model_type,
glb_sel_mdl, glb_sel_mdl_id,
glb_fin_mdl, glb_fin_mdl_id,
file=paste0(glb_out_pfx, "prdnew_dsk.RData"))
rm(submit_df, tmp_OOBobs_df)## Warning in rm(submit_df, tmp_OOBobs_df): object 'tmp_OOBobs_df' not found# tmp_replay_lst <- replay.petrisim(pn=glb_analytics_pn,
# replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
# "data.new.prediction")), flip_coord=TRUE)
# print(ggplot.petrinet(tmp_replay_lst[["pn"]]) + coord_flip())
glb_chunks_df <- myadd_chunk(glb_chunks_df, "display.session.info", major.inc=TRUE)## label step_major step_minor bgn end elapsed
## 16 predict.data.new 9 0 288.000 296.548 8.548
## 17 display.session.info 10 0 296.549 NA NANull Hypothesis (\(\sf{H_{0}}\)): mpg is not impacted by am_fctr.
The variance by am_fctr appears to be independent. #{r q1, cache=FALSE} # print(t.test(subset(cars_df, am_fctr == "automatic")$mpg, # subset(cars_df, am_fctr == "manual")$mpg, # var.equal=FALSE)$conf) # We reject the null hypothesis i.e. we have evidence to conclude that am_fctr impacts mpg (95% confidence). Manual transmission is better for miles per gallon versus automatic transmission.
## label step_major step_minor bgn end elapsed
## 11 fit.models 7 1 85.439 239.869 154.430
## 10 fit.models 7 0 35.689 85.439 49.750
## 12 fit.models 7 2 239.869 262.786 22.918
## 2 inspect.data 2 0 9.832 27.612 17.780
## 15 fit.data.training 8 1 277.062 288.000 10.938
## 14 fit.data.training 8 0 267.961 277.061 9.100
## 16 predict.data.new 9 0 288.000 296.548 8.548
## 13 fit.models 7 3 262.787 267.960 5.174
## 3 scrub.data 2 1 27.612 31.328 3.716
## 8 select.features 5 0 33.220 35.356 2.136
## 1 import.data 1 0 8.268 9.832 1.564
## 5 extract.features 3 0 31.389 32.797 1.408
## 9 partition.data.training 6 0 35.356 35.689 0.333
## 6 cluster.data 4 0 32.798 33.095 0.298
## 7 manage.missing.data 4 1 33.096 33.219 0.124
## 4 transform.data 2 2 31.328 31.388 0.060
## duration
## 11 154.430
## 10 49.750
## 12 22.917
## 2 17.780
## 15 10.938
## 14 9.100
## 16 8.548
## 13 5.173
## 3 3.716
## 8 2.136
## 1 1.564
## 5 1.408
## 9 0.333
## 6 0.297
## 7 0.123
## 4 0.060
## [1] "Total Elapsed Time: 296.548 secs"
## R version 3.2.0 (2015-04-16)
## Platform: x86_64-apple-darwin13.4.0 (64-bit)
## Running under: OS X 10.10.3 (Yosemite)
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] tcltk grid parallel stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] gdata_2.16.1 randomForest_4.6-10 arm_1.8-5
## [4] lme4_1.1-8 Matrix_1.2-1 MASS_7.3-41
## [7] rpart.plot_1.5.2 rpart_4.1-9 ROCR_1.0-7
## [10] gplots_2.17.0 dplyr_0.4.2 plyr_1.8.3
## [13] sqldf_0.4-10 RSQLite_1.0.0 DBI_0.3.1
## [16] gsubfn_0.6-6 proto_0.3-10 reshape2_1.4.1
## [19] doMC_1.3.3 iterators_1.0.7 foreach_1.4.2
## [22] doBy_4.5-13 survival_2.38-2 caret_6.0-47
## [25] ggplot2_1.0.1 lattice_0.20-31
##
## loaded via a namespace (and not attached):
## [1] Rcpp_0.11.6 class_7.3-12 gtools_3.5.0
## [4] assertthat_0.1 digest_0.6.8 R6_2.0.1
## [7] BradleyTerry2_1.0-6 chron_2.3-47 coda_0.17-1
## [10] evaluate_0.7 e1071_1.6-4 lazyeval_0.1.10
## [13] minqa_1.2.4 SparseM_1.6 car_2.0-25
## [16] nloptr_1.0.4 rmarkdown_0.7 labeling_0.3
## [19] splines_3.2.0 stringr_1.0.0 munsell_0.4.2
## [22] compiler_3.2.0 mgcv_1.8-6 htmltools_0.2.6
## [25] nnet_7.3-9 codetools_0.2-11 brglm_0.5-9
## [28] bitops_1.0-6 nlme_3.1-120 gtable_0.1.2
## [31] magrittr_1.5 formatR_1.2 scales_0.2.5
## [34] KernSmooth_2.23-14 stringi_0.5-2 RColorBrewer_1.1-2
## [37] tools_3.2.0 abind_1.4-3 pbkrtest_0.4-2
## [40] yaml_2.1.13 colorspace_1.2-6 caTools_1.17.1
## [43] knitr_1.10.5 quantreg_5.11